Request For Ex Parte Reexamination Of U.S. Patent No.

6,795,583
Attorney Docket No. 44382-0007

IN THE UNITED STATES PATENT AND TRADEMARK OFFICE

Patentee/Applicant: General Electric Company.

Inventor(s): Robert D. Barnes et al.

Patent No./App. 6,795,583 B1

No.:

Filed: November 24, 1999

Title: IMAGE DATA COMPRESSION

EMPLOYING EMBEDDED
COMPRESSION AND
DECOMPRESSION CODES

Examiner: TBD

Art Unit: TBD

Conf. No.: TBD

UNIFIED PATENTS LLC’S REQUEST FOR EX PARTE

REEXAMINATION OF U.S. PATENT NO. 6,795,583 B1

Mail Stop “Ex Parte Reexam”

Attn: Central Reexamination Unit
Commissioner for Patents
P.O. Box 1450
Alexandria, VA 22313-1450

Dear Commissioner:

Under 35 U.S.C. § 302 and 37 C.F.R. § 1.510 et seq., the undersigned, on

behalf of Unified Patents, LLC (“Unified Patents”), requests ex parte reexamination

of Claims 1 and 10 of U.S. Patent No. 6,795,583 (“the ʼ583 Patent”).

Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Based on the Office’s assignment records, the ʼ583 Patent appears to be

assigned to General Electric Company (“Patent Owner”). An assignment search

indicates that the ʼ583 Patent was assigned to the Patent Owner by an assignment

recorded at Reel 010544, Frame 0011 on January 24, 2000. No later assignments

have been found.

Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

TABLE OF CONTENTS
I. Introduction ..........................................................................................................1

II. Ex Parte Rexamination Filing Requirements ......................................................1

A. Requirements Under 37 C.F.R. § 1.510(b) ...................................................1

1. 37 C.F.R. § 1.510(b)(1) ..............................................................................1
2. 37 C.F.R. § 1.510(b)(2) ..............................................................................1
3. 37 C.F.R. § 1.510(b)(3) ..............................................................................1
4. 37 C.F.R. § 1.510(b)(4) ..............................................................................2
5. 37 C.F.R. § 1.510(b)(5) ..............................................................................2
6. 37 C.F.R. § 1.510(b)(6) ..............................................................................3
B. Fees And Deposit Account Authorization ....................................................3
C. Identification Of Other Proceedings Involving The ’583 Patent ..................3
III. Technology Background ......................................................................................3

A. Overview of Image and Video Compression, Exemplified in the Well-

Known H.263 Video Coding Standard .........................................................3
1. Overview of Prediction and Compression Algorithms, Exemplified in the
Well-Known H.263 Video Coding Standard .....................................................6
2. Overview of the Syntax for Compressed Images, Exemplified in the
Well-Known H.263 Video Coding Standard ...................................................13

IV. U.S. Patent 6,795,583 .........................................................................................21

A. Short Summary of the ’583 Patent ..............................................................21

B. Technology and Purported Invention..........................................................21
1. Image Compression Method ....................................................................22
2. Compressed Image Syntax .......................................................................25
C. The ʼ583 Patent’s Claims ............................................................................27
D. Prosecution History .....................................................................................35
E. Level of Ordinary Skill in the Art ...............................................................44

i
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

V. Claim Construction ............................................................................................44

VI. Substantial New Question of Patentability #1 – H.263+ Renders Claims 1 and

10 Obvious ...............................................................................................................45

A. Overview of Prior Art Printed Publications Raising Substantial New

Questions of Patentability ...........................................................................45
1. H.263+ (Ex. 1004) ...................................................................................45
2. Supporting Reference: Cote (Ex. 1005) ...................................................48
B. H.263+ Renders Claims 1 and 10 Obvious ................................................49
1. Proposed Rejection: H.263+ renders Claim 1 obvious ...........................50
2. Proposed Rejection: H.263+ renders Claim 10 obvious .........................72

VII. Substantial New Question of Patentability #2 – Non-Statutory

Obviousness-Type Double Patenting.......................................................................93

A. Nonstatutory Obviousness-Type Double Patenting....................................93

1. Legal Standard..........................................................................................93
2. Obviousness-Type Double Patenting in Reexamination Proceedings .....97
B. Overview of Patents Qualifying as Nonstatutory Double Patenting
References ...................................................................................................98
1. U.S. Patent 6,792,151 (Ex. 1010) ............................................................98
2. U.S. Patent 7,050,639 (Ex. 1011) ..........................................................100
C. Obviousness-Type Double Patenting raises a substantial new question of
patentability for Claims 1 and 10. .............................................................102
1. Proposed Rejection 1: The ʼ151 Patent’s Claim 10 (which also includes
the limitations of Claims 1 and 6-9) in view of H.263+ renders the ʼ583
Patent’s Claim 1 invalid based on obviousness-type double patenting .........103
2. Proposed Rejection 2: The ʼ639 Patent’s Claim 23 (which also includes
the limitations of Claim 18) in view of H.263+ renders the ʼ583 Patent’s
Claim 10 invalid based on obviousness-type double patenting.....................112

VIII. Conclusion.................................................................................................122

ii
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

I. INTRODUCTION
Unified Patents, LLC (“Unified Patents”) respectfully requests ex parte

reexamination of Claims 1 and 10 of U.S. Patent 6,795,583 (the “’583 Patent”)

(EX1001) and issuance of a reexamination certificate canceling Claims 1 and 10.

II. EX PARTE REXAMINATION FILING REQUIREMENTS

A. Requirements Under 37 C.F.R. § 1.510(b)

1. 37 C.F.R. § 1.510(b)(1)
Statements pointing out at least one substantial new question of patentability

based on prior art patents and printed publications for Claims 1 and 10 of the ʼ583

Patent are provided in this Request.

2. 37 C.F.R. § 1.510(b)(2)
Reexamination of Claims 1 and 10 of the ʼ583 Patent is requested, and a

detailed explanation of the pertinence and manner of applying the cited prior art to

every claim for which reexamination is requested is provided in this Request.

3. 37 C.F.R. § 1.510(b)(3)
Copies of every patent or printed publication relied upon or referred to in the

statement pointing out each substantial new question of patentability or in the

detailed explanation of the pertinence and manner of applying the cited prior art are

provided as Exhibits 1002-1014 to this Request. Attachment 1 is an information

disclosure statement citing those references.

1
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

4. 37 C.F.R. § 1.510(b)(4)
A copy of the ʼ583 Patent is provided as Exhibit 1001 to this Request.

5. 37 C.F.R. § 1.510(b)(5)
The attached Certificate of Service (Attachment 2) indicates that a copy of

this Request, in its entirety, has been served on the apparent Patent Owner at the

correspondence address for the ’583 Patent indicated in PAIR, in accordance with

37 C.F.R. § 1.33(c):

GE Healthcare
c/o Fletcher Yoder, PC
P.O. Box 692289
Houston, Texas 77269-2289

The attached Certificate of Service (Attachment 2) also indicates that a

courtesy copy of this Request, in its entirety, has also been served on GE Video

Compression, LLC at the correspondence address indicated in PAIR for U.S. Patent

No. 10,057,603, which is assigned to GE Video Compression, LLC:

GE Video Compression, LLC

c/o Morgan, Lewis & Bockius LLP (WA) GE 128571
1111 Pennsylvania Avenue, N.W.
Washington, DC 20004

2
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

6. 37 C.F.R. § 1.510(b)(6)
Unified Patents certifies that the statutory estoppel provisions of 35 U.S.C.

§ 315(e)(1) and 35 U.S.C. § 325(e)(1) do not prohibit Unified Patents from filing

this ex parte patent reexamination request.

B. Fees And Deposit Account Authorization

Pursuant to 37 C.F.R. § 1.20(c)(2), Unified Patents is concurrently submitting

$12,600 for the reexamination filing fee. The Office is authorized to charge any

deficiency, or credit any overpayment, to Deposit Account No. 50-6990.

C. Identification Of Other Proceedings Involving The ’583 Patent

As of the filing date of this request, Unified Patents is not aware of any other

proceedings involving the ʼ583 Patent.

III. TECHNOLOGY BACKGROUND

A. Overview of Image and Video Compression, Exemplified in the

Well-Known H.263 Video Coding Standard
As illustrated in Figure 2.2 of EX1006 below, digital images are made up of

an array of pixels. (EX1002 at ¶26; EX1006 at Figure 2.2.) In turn, video data is

made up of a temporal series of images, where the images are successively displayed

at a certain rate to render video for display. (Id.)

3
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(EX1006, Figure 2.2.1)

While digital images are subdivided in their lowest layer as pixels, images are

often also subdivided into multiple intervening layers defined by iteratively smaller

subregions. (EX1002 at ¶27.) For example, as illustrated below, in the well-known

H.263 video coding standard, the coded images (termed “pictures” in H.263) are

defined in iteratively smaller subregions called groups of blocks (GOBs),

macroblocks, and blocks. (EX1002 at ¶27; “H.263+: Video Coding at Low Bit

Rates,” Guy Cote et al, IEEE Nov. 1998 (“Cote” (EX1005)) at Figure 2). To note,

Figure 2 below uses the term “pel”—“pel” is synonymous with “pixel.” (EX1002

at ¶27.)

1
In this Request, all annotations and emphasis are added unless indicated otherwise.

4
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(Cote at Figure 2.)

Specifically, each macroblock includes six blocks—four luminance blocks

and two chroma blocks, where these blocks generally describe different aspects of

the pixels’ color/intensity. (EX1002 at ¶28; see also ITU-T Recommendation

H.263, “Video coding for Low Bit Rate Communication,” published in 1998, 1-2

(“H.263+” (EX1004)) at 11 (describing macroblocks as “consist[ing] of four

luminance blocks and the two spatially corresponding colour difference [i.e. chroma]

5
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

blocks”), Figure 5 (Page 11) (showing the “arrangement of blocks in a

macroblock”).) Each block then comprises the underlying array of pixels. (EX1002

at ¶28 (noting that Figure 2 of Cote shown above is merely illustrative and

pictures/blocks can include any number of pixels); see also H.263+ at 7 (describing

the H.263 picture format as being “determined by [a variable] number of pixels per

line”).)

1. Overview of Prediction and Compression Algorithms,

Exemplified in the Well-Known H.263 Video Coding Standard
Digital image and video data requires substantial computer memory when

stored and substantial bandwidth when transmitted. (EX1002 at ¶29.) To account

for this, image and video data can be compressed for storage and transmission. (Id.)

To compress image data, coders rely on prediction and compression algorithms.

(Id.) To apply these algorithms, image and video coders rely on coding standards

that define the prediction and compression algorithms that should be used and the

syntax for the coded image/video data. (Id.) For example, the JPEG standard, which

was first published in 1992, is popular for compressing still images. (Id.) The H.263

standard, which was first published in 1996, is popular for video coding. (EX1002,

¶29; see also H.263+ (including the second version of the H.263 standard, which

was published in 1998 and is commonly referred to as H.263+ or H.263v2); Cote

(describing the H.263+ standard).)

6
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

At a high level, prediction algorithms predict data to be coded based on the

content of other data. (EX1002 at ¶30.) As a generalized example, the values for a

subregion (such as a macroblock) of an image depicting a clear blue sky are used as

the predicted values for an adjacent subregion depicting the same clear blue sky.

(Id.) To make a prediction, the coder calculates the difference between the

subregion currently being coded and what was predicted (i.e., the adjacent,

previously coded subregion). (Id.) If the prediction is accurate, the difference

calculated would be close to zero, minimizing the amount of data required to code

the subregion. (Id.)

Moreover, in H.263, the terms “inter” and “intra” are used to describe

different types of prediction. (EX1002 at ¶31.) These terms are used to describe

data in both the picture layer and macroblock layer. (Id.) In the picture layer, inter-

pictures are called P-pictures, where the P-picture’s macroblocks are predicted based

on macroblocks in a temporally previous picture. (H.263+ at 11 (“INTER: A picture

using a temporally previous reference picture (also called a P-picture).”); EX1002 at

¶31; Cote at 849-852.) Intra-pictures, on the other hand, are called I-pictures and do

not rely on temporally proximate pictures for prediction. (H.263+ at 11 (“INTRA:

A picture having no reference picture(s) for prediction (also called an I-picture.”));

EX1002 at ¶31; Cote at 849-852.) For example, with the “Advanced Intra Coding

mode” of H.263+, I-picture’s macroblocks are predicted based on macroblocks

7
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

within the same picture. (H.263+ at 4 (in the Advanced Intra Coding mode, “intra

blocks are first predicted from neighboring intra blocks [within the same picture]

prior to coding”); EX1002 at ¶31.)

Thus, as illustrated in Figure 7a of Cote below, H.263+ describes P-pictures

(inter) pictures that are predicted based on a temporally previous pictures and I-

pictures (intra) that do not rely on other pictures for prediction. (Cote at 854-855,

Figure 7(a); EX1002 at ¶32.) In Cote’s Figure 7a, the arrows indicate the direction

of prediction, where P-pictures rely on a single temporally previous picture. (Cote

at 854-855, Figure 7(a); EX1002 at ¶32.)2

(Cote at Figure 7(a).)

2
Cote’s Figure 7(a) also describes “B” pictures, which rely on both temporally

previous and subsequent pictures. (Cote at 854-855, Figure 7(a); see also H.263+

at 12 (“B: A picture having two reference pictures, one of which temporally precedes

the B-picture and one of which temporally succeeds the B-picture and has the same

picture size (see Annex O)”).)

8
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

As described above, it was well-known that the applied prediction algorithm

(e.g., I-picture and P-picture) can be defined in the picture layer. (EX1002 at ¶33.)

Additionally, H.263 also defines prediction type (intra and inter) in the macroblock

layer. (Id.) For I-pictures and P-pictures, H.263+ defines multiple types of inter

and intra macroblocks in Table 9:

(H.263+ at Table 9 (Page 36).)

As seen in Table 9, for “INTER” picture type (P-pictures)—as indicated in the left

column—different types of inter and intra macroblocks can be used. (EX1002, ¶34;

H.263+ at Table 9 (Page 36) (providing that INTER pictures (P-pictures) can rely

on “MB type[s]” 0-5, where these “MB type[s]” are called “INTER,” “INTER+Q,”

“INTER4V,” “INTRA,” “INTRA+Q,” and “INTER4V+Q”).) And for “INTRA”

Picture type (I-picture), only intra macroblocks can be used. (EX1002, ¶34; H.263+

9
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

at Table 9 (Page 36) (providing that INTRA pictures (I-pictures) can rely on “MB

type[s]” 3-4, where these “MB type[s]” are called “INTRA,” and “INTRA+Q”).)

Thus, I-pictures do not include inter macroblocks and, instead, only include

intra macroblocks (MB Types 3-4). (H.263+ at 34-36; EX1002 at ¶35.) On the other

hand, P-pictures can include both intra and inter macroblocks (MB Types 0-5).

(H.263+ at 34-36; EX1002 at ¶35.) In that way, a P-picture can include macroblocks

that are predicted based on corresponding macroblocks in a nearby pictures (inter

macroblocks), while other macroblocks within the same P-picture are not (intra

macroblocks). (H.263+ at 11-12 (“The coding mode in which temporal prediction

is applied is called INTER; the coding mode is called INTRA if no temporal

prediction is applied. The INTRA coding mode can be signalled [sic] at the picture

level (INTRA for I-pictures or INTER for P-pictures) or at the macroblock level in

P-pictures.”), 73 (for the Advanced Intra Coding mode, describing that macroblock

prediction is “signalled [sic] on a macroblock-by-macroblock basis”); EX1002 at

¶35; Cote at 852 (“The two switches in Fig. 1 represent the intra/inter mode

selection, which is not specified in the standard. Such a selection is made at the

macroblock level.”).)

Moreover, at a high level, compression algorithms, such as entropy coding

schemes, look for redundancies in a data set to take advantage of to minimize the

number of bits used for coding. (EX1002 at ¶36; see Cote at 851 (describing the use

10
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

of entropy coding in H.263).) As a generalized example, if an image is almost

entirely blue, then the compression algorithm could encode the image using a

scheme that relies on less bits to code blue pixels. (EX1002 at ¶36.) To do so,

compression algorithms rely on code tables (often referred to as “Variable Length

Code (VLC)” tables, “compression code tables,” etc.) that align original bit values

with compressed bit values to which they are translated. (Id.) These compression

code tables are stored by the coder and identified in the compressed data so that the

same tables are relied on for coding and decoding. (Id.) Indeed, after such

compression, the coder would be unable to decompress the coded data without

knowing which compression code table was relied on for coding. (Id.)

For example, the well-known H.263 standard defines VLC tables that are used

to entropy code (apply a compression algorithm to) predicted macroblocks.

(EX1002 at ¶37; H.263+ at 11 (“the [predicted] coefficients are quantized and

entropy coded”).) In standard H.263 operation (such as occurs when the Advanced

Intra Coding mode is not applied), “the standard employs the same VLC table for

coding all quantized coefficients.” (Cote at 854; EX1002 at ¶37.)

Yet, it was well-known that H.263+ could also select different compression

algorithms (VLC tables) to apply to compress macroblocks. For example, H.263+

describes an Advanced Intra Coding mode in Annex I that is a selectable mode that

“alters [encoding and] decoding of macroblocks of type ‘INTRA’ (macroblocks of

11
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

other types are not affected).” (H.263+ at 73-80, 73; EX1002 at ¶38.) The

Advanced Intra Coding modes alterations to the coding of intra macroblocks

includes the reliance on a “separate VLC” table for intra macroblocks. (H.263+ at

73 (“The coding efficiency of INTRA macroblocks is improved by using: … a

separate VLC for INTRA coefficients”).) Thus, in the Advanced Intra Coding mode,

the standard chooses between two VLC tables—one for intra macroblocks and one

for inter macroblocks. (EX1002 at ¶38; H.263+ at 4 (“Separate Variable Length

Code (VLC) tables are defined for the INTRA blocks”), 74-75).)

Further, it was known that relying on different compression algorithms for

different macroblock types (intra or inter) takes advantage of known differences in

the predicted data. (EX1002 at ¶39; Cote at 854 (“The main part of the [H.263+]

standard employs the same VLC table for coding all quantized coefficients.

However, this table is designed for inter macroblocks and is not very effective for

coding intra macroblocks.”), 854 (for the Advanced Intra Coding mode, explaining

that a different VLC table is used for intra macroblocks that “is optimized to global

statistics of intra macroblocks” where “larger coefficients with smaller runs of zeros

are more common [than in inter macroblocks]”).) In that way, when the Advanced

Intra Coding mode is applied, it was known that H.263+ selects different

compression algorithms based on macroblock type (intra or inter). (Id.)

12
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

2. Overview of the Syntax for Compressed Images, Exemplified

in the Well-Known H.263 Video Coding Standard
As discussed above, to increase efficiency, coding standards such as H.263+

select the prediction and compression algorithms to be used for coding image data.

(EX1002 at ¶40.) After prediction and compression algorithms are selected, they

must be identified in the compressed image file for re-use during decoding. (Id.) To

do so, selected prediction and compression algorithms, and other aspects of

compressed image data, are identified in headers. (Id.)

For example, in H.263, the Picture, GOB, and Macroblock layers include

headers, and in these headers, different aspects of the compressed image file,

including which prediction and compression algorithms were used, are identified.

(EX1002 at ¶41; H.263+ at 22, 32-34.) As illustrated below, H.263+ provides

detailed block diagrams for the syntax of the headers used for each picture, GOB,

and macroblock. (H.263+ at Figures 7 (Page 22), 9 (Page 32), and 10 (Page 34).)

13
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at Figures 7 (Page 22), 9 (Page 32), and 10 (Page 34).)

Specifically, as illustrated above, H.263+ explains that the picture header

includes a “PTYPE” syntax element that identifies the selected “Picture Coding

Type” as an “INTRA (I-picture)” or “INTER (P-picture).” (H.263+ at 23; EX1002

at ¶42.) H.263+ further explains that, for coded macroblocks 3 , the macroblock

header includes a “MCBPC” syntax element which defines the “macroblock type,”

3
Macroblocks can also be uncoded. (EX1002, at ¶42; see also H.263+ at 34

(“MCBPC is always included in coded macroblocks.”), 36 (Table 9 showing “Not

coded” macroblocks as a macroblock type); Cote at 852 (“If a macroblock does not

change significantly with respect to the reference picture, an encoder can also choose

not to encode it, and the decoder will simply repeat the macroblock located at the

subject macroblock’s spatial location in the reference picture.”).)

14
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

including whether the macroblock is intra or inter. (H.263+, 33-36, 74 (“MCBPC

[may] indicate[] a macroblock of type INTRA”); EX1002, at ¶42.) I-pictures—

where picture type is indicated in the PTYPE element of the picture header—include

only intra macroblocks as defined on a macroblock-basis in the MCBPC element of

the macroblock header. (EX1002, at ¶42; H.263+ at 23, 33-36 (describing multiple

types of intra macroblocks applicable to I-pictures in Tables 7 and 9).) P-pictures

may include both inter and intra macroblocks as defined on a macroblock-basis in

the MCBPC element of the macroblock header. (EX1002, at ¶42; H.263+ at 23, 33-

36.)

Additionally, as illustrated below, H.263+ explains that the picture header can

include additional syntax elements, such as the “PLUSPTYPE” element. (EX1002,

¶43; H.263+ at Figures 7, 8 (Page 22).) The PLUSPTYPE element is an optional

element that is present only if indicated by the PTYPE element. (EX1002, ¶43;

H.263+ at 24 (describing the PLUSPTYPE element as “present only if [its] presence

… is indicated in bits 6-8 of PTYPE.”).) When the PLUSPTYPE element is present,

the PLUSPTYPE element identifies picture type instead of the PTYPE element.

(EX1002, ¶43; H.263+ at 23 (explaining that if the PTYPE only identifies the

“Picture Coding Type” in “Bit 9” when the PTYPE also indicates that the

PLUSPTYPE element is not present (“If bits 6-8 are not equal to ‘111’”).) While

the PTYPE element is limited to identifying a picture as an “INTRA (I-picture)” or

15
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

“INTER (P-picture),” the PLUSPTYPE element’s MPPTYPE subfield identifies

between more picture types, including “I-picture (INTRA),” “P-picture (INTER),”

“B-picture,” and various others. (EX1002, ¶43; H.263+ at 25 (providing that the

MPPTYPE subfield of the PLUSPTYPE element identifies “Picture Type” in “Bits

1-3”).)

(H.263+ at Figures 7, 8 (Page 22) (annotated to include MPPTYPE and OPPTYPE

subfields).)

As illustrated above, the “PLUSPTYPE” field in an OPPTYPE subfield identifies

whether optional prediction modes, such as the Advanced Intra Coding mode, are

being applied. (EX1002, ¶44; H.263+ at 24 (stating that “Bit 8” of the OPPTYPE

subfield of PLUSPTYPE element identifies whether the “Optional Advanced Intra

16
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Coding (AIC) mode” is being used), 73 (“The use of [the Advanced Intra Coding

mode] is indicated in the PLUSPTYPE field of the picture header.”).)

And when the PLUSPTYPE field in the picture header identifies the

Advanced Intra Coding mode’s use, H.263+ describes a modified syntax for the

macroblock header including an INTRA_MODE element, as illustrated below.

(EX1002, ¶45; H.263+ at 73-74, Figure I.1.) “INTRA_MODE is present only when

the MCBPC indicates a macroblock of type INTRA (macroblock type 3 or 4).”

(H.263+ at 74.) The “INTRA_MODE” field specifies aspects of the advanced intra

prediction algorithms being applied to the intra macroblock—such as whether the

advanced intra prediction algorithm should rely on neighboring blocks to the left or

above the block being predicted. (EX1002, ¶45; H.263+ at 73-75, 75 (“Prediction

mode = 1 uses the vertically adjacent block in forming a prediction. … Prediction

mode = 2 uses the horizontally adjacent clock in forming a prediction.”).)

17
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at Figure 10, I.1.)

Additionally, if the OPPTYPE subfield of the PLUSPTYPE element in the

picture header identifies the use of the Advanced Intra Coding mode, macroblocks

that are identified as intra-type in the MCBPC element of the macroblock header

rely on a different VLC table (and thereby a different compression algorithm), as

described in Section III.A.1 above. (EX1002 at ¶46; H.263+ at 4 (“Separate

Variable Length Code (VLC) tables are defined for the INTRA blocks”), 74-75

(using a “separate VLC for INTRA coefficients”).)

Thus, it was known from coding standards such as H.263+ how to fine-tune

the coding of image data using selected prediction and compression algorithms.

(EX1002 at ¶¶40-47.) For example, the figure below illustrates the syntax of

compressed data when the Advanced Intra Coding mode is applied. As explained

18
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

above and summarized below, in these circumstances, the MPPTYPE subfield of the

PLUSPTYPE element in the picture header indicates a selectable prediction

algorithm, and the MCBPC element in the macroblock header indicates a selectable

compression algorithm. (EX1002 at ¶47.)

(H.263+ at Figures 7 (Page 22), 8 (Page 22) (annotated to include MPPTYPE and

OPPTYPE subfields), 9 (Page 32), and I.1 (Page 74).)

19
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

And to summarize, when the application of the Advanced Intra Coding mode

is indicated in the OPPTYPE subfield of the PLUSPTYPE element in the picture

header, the encoder/decoder compresses macroblocks according to two compression

code options (two VLC tables) based on macroblock type—intra or inter—as

determined by the MCBPC element of the macroblock header. (EX1002 at ¶48.)

And when the PLUSPTYPE element in the picture header is present in the syntax of

the encoded data (where it must be present if it’s indicating that the Advanced Intra

Coding mode is activated), the MPPTYPE subfield of the PLUSPTYPE element

(instead of the PTYPE element) indicates the selectable prediction algorithm

between at least I-picture (Intra) and P-picture (Inter). (EX1002 at ¶49.)

It was well-known that, with the Advanced Intra Coding mode of H.263+, the

selection of prediction and compression algorithms improves performance by taking

advantage of redundancies in the image data. (EX1002 at ¶50; Cote at 853-854

(explaining that the Advanced Intra Coding mode “improves compression

performance when coding intra macroblocks [because] [i]n this mode, inter block

prediction from neighboring intra coded blocks [and] a separate VLC table for intra

coded coefficients are employed” and where the VLC table employed “is optimized

to global statistics of intra macroblocks”.)

20
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

IV. U.S. PATENT 6,795,583

A. Short Summary of the ’583 Patent

The ʼ583 Patent describes selecting prediction and compression algorithms

for regions and subregions, respectively, of an image data stream. (EX1002 at ¶51;

ʼ583 Patent at 2:3-28, 14:2-22, 14:50-61, Figure 6.) The ʼ583 Patent purports that

selecting algorithms optimizes image compression. (EX1002 at ¶51; ʼ583 Patent at

1:54-63.) The ʼ583 Patent further describes the syntax for the coded images

where, in the syntax, the selected prediction and compression algorithms are

identified. (EX1002 at ¶51; ʼ583 Patent at Figure 13.)

B. Technology and Purported Invention

The ʼ583 Patent generally presents steps for compressing image data (ʼ583

Patent at Figure 6) and a syntax for organizing the compressed image data (ʼ583

Patent at Figure 13). (EX1002 at ¶52.)

Although the ʼ583 purports to present an improvement to image compression

techniques, it merely discloses known information. For example, as demonstrated

in the side-by-side figures below, the ʼ583 Patent’s Figure 13 (described in further

detail in Section VI.B.2 below) describes prediction algorithm selection and

compression algorithm selection, where these concepts were already known from

H.263+ (as described in Section III above) as demonstrated in the use of at least

21
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

the PLUSPTYPE element (for prediction algorithm selection) and MCBPC field (for

compression algorithm) under the Advanced Intra Coding mode. (EX1002 at ¶53.)

ʼ583 Patent: Figure 13 H.263+: Figures 7, 8, 9, and I.1

1. Image Compression Method

As seen in Figure 6 below, the ʼ583 Patent’s compression of image data boils

down to two key steps: (1) image prediction and (2) compression of image

subregions. For each of these steps, the ʼ583 Patent describes the selection of

algorithms that are relied upon. (EX1002 at ¶54.)

For prediction and referring to Figure 6 below, the ʼ583 Patent provides that

after an image is received (step 276), image descriptions are considered (steps 278

and 280), and a prediction algorithm is selected based on the image descriptors (step

22
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

282). (ʼ583 Patent at Figure 6.) The prediction algorithm can be selected based on

descriptive information for the image such as “the modality of the originating image

system, the study type or anatomy featured in the image, the number of columns in

the image, the number of rows, … the computational efficiency desired, the

processing power of the system, and so forth[.]” (ʼ583 Patent at 10:59-11:3; EX1002

at ¶55.)

23
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(ʼ583 Patent at Figure 6.)

As an example, the ʼ583 Patent provides that a selected prediction algorithm

can predict pixel values by calculating the difference between neighboring pixels.

(ʼ583 Patent at 9:38-50 (“[O]ne or more of several predictor algorithms may be

employed … [including] a simple and straightforward algorithm [where] the value

of each pixel p(i, j) is predicted to be the value of the immediately preceding pixel

p(i-1, j). … The resulting difference values form a matrix of equal size to the original

image matrix.”); EX1002 at ¶56.)

For compression and as seen in steps 274 and 284 of Figure 6 above, the ʼ583

Patent explains that images can be subdivided into subregions. (ʼ583 Patent at 11:3-

9.) The ʼ583 Patent then provides that optimum compression tables are selected

(step 292) to compress the image subregions (step 296). (ʼ583 Patent at Figure 6.)

The ʼ583 Patent explains that optimum compression tables—where, as explained in

Section III.A above, the compression tables relied upon correspond to compression

algorithms—can be selected “based upon certain image characteristics [such as]

specific image types originating in specific modalities, such as CT or MR images,

[where CT or MR images] may be best compressed with specific candidates

tables[.]” (ʼ583 Patent at 11:10-18.) Additionally, while the ʼ583 Patent presents

the selected prediction algorithm as being applicable to the entire image, the ʼ583

Patent provides the selected compression tables as being applicable to individual

24
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

subregions. (ʼ583 Patent at 7:1-3 (“the subregions may be optimally compressed in

accordance with one of several candidate compression algorithms”); EX1002 at ¶57.)

2. Compressed Image Syntax

The ʼ583 Patent also describes that the compressed image file can be defined

by a certain syntax. This syntax is provided in the ʼ583 Patent’s Figure 13 below.

Particularly, the ʼ583 Patent provides that the image can include a compression

header (item 198) which identifies the selected prediction algorithm (item 207).

(ʼ583 Patent at 13:5-10.) Figure 13 further shows that the compressed subregions

(202, 204) include a key code identifying the compression tables applied (214, 218)

and the compression code (216, 220). (ʼ583 Patent at 13:21-27; EX1002 at ¶58.)

(EX1001, Figure 13)

25
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Further, the ʼ583 Patent provides an embodiment where the compression

header can also include an identification segment (210) “to identify which

compression algorithms (e.g. compression code tables) were selected for the

compression, and the manner in which each table is encoded in the data stream for

each subregion.” (ʼ583 Patent at 13:9-13.) For example, if ten compression tables

are available for use at a coder, Figure 6’s identification segment embodies tables 0,

3, 4, and 5 (as seen in the bottom row of the identification segment) being selected

for compression, where each of these identified tables (0, 3, 4, and 5) is aligned with

a two bit sequence used in the subregion’s key code. (ʼ583 Patent at 13:14-18.) For

example, key code 214 with the value “11” aligns to compression code table “0”

based on the embodied identification segment. (See also ʼ583 Patent at 5:34-50

(regarding the embodiment of Figure 2, explaining that “additional data is stored

within the compressed image data, cross-referencing the algorithms identified in

compression header 60 for use in decompressing the image data. Specifically, in a

presently preferred embodiment, the compression header 60 includes … references

to specific optimal algorithms, in the form of compression code tables used to

compress the subregions optimally.”); EX1002 at ¶59.)

26
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

C. The ʼ583 Patent’s Claims

This Request challenges Claims 1 and 10 of the ’583 Patent. Below, Tables

1 and 2 list portions of the specification that provide exemplary support for the claim

limitations.

Table 1

Claim 1 limitations Non-limiting Examples from the

Specification
1. A method for compressing an “The preferred technique for compressing
image data stream, the the image data stream in the system
method comprising the steps of: described above to create hybrid
compressed image data files is summarized
in FIG. 6.” (ʼ583 Patent at 7:31-33.)

(a) dividing the image data stream “At step 284, the subregion size for
into a plurality of subregions, division of the image data stream into
subregions is selected in accordance with
the preferences established at step 274.
Again, step 284 may consist of a default
selection, which may be altered depending
upon some or all of the characteristics or
factors considered for selection of the
predictors.” (ʼ583 Patent at 11:3-9; see
also ʼ583 Patent at Figure 6.)

“As described below, this compression

header will contain code identifying the
version of the compression routine, the
predictors selected at step 282, the
subregion sizes selected at step 284, and so
forth.” (ʼ583 Patent at 11:18-22.)

“FIG. 13 represents an image data set

compressed in accordance with the
foregoing technique. The image data set,

27
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

designated generally by the reference

numeral 194, includes the descriptive
header 196 appended at step 160, along
with the compression header 198 affixed at
step 192. Following the compression
header 198 is the compressed image data
200 which includes compressed image
subregions 202, 204 and so forth. Each
compressed data subregion, in turn,
includes an identification of the algorithm
or algorithms (e.g. compression code
tables) used to encode the subregion,
followed by the actual compressed data.”
(ʼ583 Patent at 12:60-13:3; see also ʼ583
Patent at Figure 13.)

(b) selecting a predictor algorithm “Based upon the preferences set in the
from a plurality of predictor configuration segment 272, predictors are
algorithms to generate predicted selected at step 282 depending upon the
pixel data values … image characteristics identified as step 280.
Again, these may include the modality of
the originating imaging system, the study
type or anatomy featured in the image, the
number of columns in the image, the
number of rows, and so forth. Moreover,
other factors may be considered in
selecting the predictors at step 282, such as
the computational efficiency desired, the
processing power of the system, and so
forth, with computationally efficient
predictors being selected where such
processor capabilities are limited, or where
additional speed is desired.” (ʼ583 Patent
at 10:59-11:3.)

“At compression segment 256 of the

control logic, a set of prediction errors or
difference values are computed at step 290.
As noted above, these values are based

28
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

upon application of one or more of the

predictor algorithms selected at step 282,
and by subsequently comparing the
predicted values with the actual values for
each pixel to determine the prediction error
or difference. The resulting prediction
errors then form a data stream with the first
pixel being assigned its actual value,
followed by the difference values for each
pixel of the subregion.” (ʼ583 Patent,
11:23-32.)

and compressing data of each “Evaluation segment 254 continues with

subregion in accordance with one the selection of a subset of compression
of a plurality of compression tables, where appropriate, as indicated at
algorithms; step 286. In particular, based upon certain
image characteristics, it may be useful to
preselect certain compression tables as
defaults. For example, specific image types
originating in specific modalities, such as
CT or MR images, may be best compressed
with specific candidate tables which may
be selected at step 286. At step 288 a
compression header is affixed to the image
data. As described below, this compression
header will contain code identifying the
version of the compression routine, the
predictors selected at step 282, the
subregion sizes selected at step 284, and so
forth.” (ʼ583 Patent, 11:10-22.)

“At step 296, the image data for the

subregion is compressed by application of
the selected compression code table. The
series of steps of segment 256 is repeated
for each subregion of the image until the
entire image is compressed.” (ʼ583 Patent,
12:21-25.)

29
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(c) compiling a compressed data “FIG. 13 also illustrates the format in the
file including compressed data for presently preferred embodiment for the
each subregion and key code compressed data code for each image
identifying the compression subregion 202, 204, and so forth. In the
algorithms applied to compress illustrated embodiment, a first portion of
each subregion, … each subregion includes an identification of
the compression table used to compress the
subregion data, as indicated at reference
numeral 214 for subregion 202, and
reference numeral 218 for subregion 204.
This key code is followed by the
compression codes as indicated at reference
numerals 216 and 220, respectively.
Finally, FIG. 13 illustrates the code
inserted at the end of each compressed
image file, or portion of a file. Specifically,
following the last subregion BN of the
image, the end of block code portion 222 is
affixed. As noted above, this end of block
code signals the end of an image, or may
be used to signal the end of a portion of an
image, where the data compression routine
is changed within a single image, such as
due to large variations in the image data
entropy. The padding code is inserted as
indicated at reference numeral 224. This
code may be of variable size as needed to
complete the compressed data file on a
whole word length. Finally, a 32 bit
checksum portion 226 is added to complete
the compressed data.” (ʼ583 Patent at
13:19-39; see also ʼ583 Patent at Figure
13.)

wherein the compressed data “Each image is comprised of a matrix

comprises pixel data that includes having a width 104 and a height 106
predictor error data representative defined by the number and distribution of
of differences between predicted individual pixels 108. The pixels of the
image matrix are arranged in rows 110 and

30
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

pixel data values and actual pixel columns 112, and will have varying
data values, and characteristics which, when viewed in the
reconstructed image, define the features of
interest.” (ʼ583 Patent at 5:61-66; see also
ʼ583 Patent at Figure 3.)

“In the presently preferred embodiment,

one or more of several predictor algorithms
may be employed. Referring again to FIG.
5, in a simple and straightforward
algorithm, the value of each pixel p(i, j) is
predicted to be the value of the
immediately preceding pixel p(i-1, j). This
predictor algorithm provides a
computationally extremely efficient tool
for the prediction of each pixel value, with
the first pixel in the image being estimated
at a value of zero. The difference value is
then generated by finding the absolute
value of the difference between the
predicted value and the actual value for the
pixel of interest. The resulting difference
values form a matrix of equal size to the
original image matrix.” (ʼ583 Patent at
9:38-50; see also ʼ583 Patent at Figure 5.)

(d) appending a compression “FIG. 13 represents an image data set

header to the compressed data file, compressed in accordance with the
the compression header including foregoing technique. The image data set,
predictor data identifying the designated generally by the reference
selected predictor algorithm used numeral 194, includes the descriptive
to generate the predicted pixel data header 196 appended at step 160, along
values. with the compression header 198 affixed at
step 192. Following the compression
header 198 is the compressed image data
200 which includes compressed image
subregions 202, 204 and so forth. Each
compressed data subregion, in turn,
includes an identification of the algorithm

31
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

or algorithms (e.g. compression code

tables) used to encode the subregion,
followed by the actual compressed data.”
(ʼ583 Patent at 12:60-13:3; see also ʼ583
Patent at Figure 13.)

“FIG. 13 also illustrates a presently

contemplated format for a compression
header 198. As shown in FIG. 13, the
compression header includes an
identification of the compression routine
version 206, followed by an identification
of the predictors used in the compression
process 207, and identification of the
subregion length at reference numeral 208.
A compression algorithm identification
segment 210 is then inserted in the
compression header to identify which of
the compression algorithms (e.g.
compression code tables) were selected for
the compression, and the manner in which
each table is encoded in the data stream for
each subregion. In the example of FIG. 13,
for example, tables 0, 3, 4 and 5 were
selected, as indicated at reference numeral
212, with each table being identified by a 2
bit binary code to be found within the first
2 bits of the compressed data for each
subregion.” (ʼ583 Patent at 13:4-18.)

Table 2

Claim 10 limitation Specification Support

10. A compressed image data file “FIG. 13 represents an image data set
comprising: compressed in accordance with the
foregoing technique.” (ʼ583 Patent at

32
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

12:60-61; see also ʼ583 Patent at Figure

13.)

a plurality of compressed data “FIG. 13 also illustrates the format in the

segments, each segment including presently preferred embodiment for the
compressed data code for a compressed data code for each image
subregion of an image data stream; subregion 202, 204, and so forth. In the
illustrated embodiment, a first portion of
each subregion includes an identification of
the compression table used to compress the
subregion data, as indicated at reference
numeral 214 for subregion 202, and
reference numeral 218 for subregion 204.
This key code is followed by the
compression codes as indicated at reference
numerals 216 and 220, respectively.”
(ʼ583 Patent at 13:19-27; see also ʼ583
Patent at Figure 13.)

a plurality of compression key code “FIG. 13 also illustrates the format in the
segments, each key code segment presently preferred embodiment for the
being associated with at least one compressed data code for each image
compressed data segment and subregion 202, 204, and so forth. In the
identifying a compression illustrated embodiment, a first portion of
algorithm employed to generate the each subregion includes an identification of
respective compressed data code the compression table used to compress the
from data for the subregion of the subregion data, as indicated at reference
image data stream; numeral 214 for subregion 202, and
reference numeral 218 for subregion 204.
This key code is followed by the
compression codes as indicated at reference
numerals 216 and 220, respectively.”
(ʼ583 Patent at 13:19-27; see also ʼ583
Patent at Figure 13.)

a compression header including a “At step 294 key code for the subregion
cross reference of the key code compression is inserted into the
segments to the compression compressed data stream, immediately
algorithms. preceding the compressed subregion data.

33
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

As noted above, where four candidate

tables are employed for the compression,
the code inserted at step 294 may include a
pair of designated bits. The compression
header created at step 288 cross references
this key code to the selected compression
tables. At step 296, the image data for the
subregion is compressed by application of
the selected compression code table.” (ʼ583
Patent at 12:14-23.)

“FIG. 13 also illustrates a presently

contemplated format for a compression
header 198. As shown in FIG. 13, the
compression header includes an
identification of the compression routine
version 206, followed by an identification
of the predictors used in the compression
process 207, and identification of the
subregion length at reference numeral 208.
A compression algorithm identification
segment 210 is then inserted in the
compression header to identify which of
the compression algorithms (e.g.
compression code tables) were selected for
the compression, and the manner in which
each table is encoded in the data stream for
each subregion. In the example of FIG. 13,
for example, tables 0, 3, 4 and 5 were
selected, as indicated at reference numeral
212, with each table being identified by a 2
bit binary code to be found within the first
2 bits of the compressed data for each
subregion.” (ʼ583 Patent at 13:4-18.)

34
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

D. Prosecution History
The ʼ583 Patent began as U.S. Patent App. No. 09/448,938 (“the ʼ938

Application”), which was filed November 24, 1999. (EX1001, Cover Page.) The

ʼ938 Application originally included 28 claims, including original claims 1 and 12,

which became Claims 1 and 10 of the ʼ583 Patent, respectively. (EX1003 at 71-72.)

Original claims 1 and 12 generally provided for the compression of image

subregions using selected compression algorithms. (EX1003 at 71-72.)

Additionally, original claims 5 and 6 required using prediction algorithms and

storing a selected prediction algorithm in a header (as the Challenged Claim 1 of the

‘583 Patent now requires). (EX1003 at 71-72.)

In the first office action, the Examiner rejected the claims of the ʼ938

Application, including original claims 1 and 12, under 35 U.S.C. § 102(a) as

anticipated by U.S. Patent No. 5,845,088 (“Lewis,” EX1007). (EX1003 at 122-124.)

The Examiner additionally rejected dependent claims 5 and 6 under 35 U.S.C.

§ 103(a) as obvious over Lewis in view of U.S. Patent No. 5,764,374 (“Seroussi,”

EX1008). (EX1003 at 124-125.) Without amending the claims, the Applicant

responded to the first office action with arguments that the Lewis reference

“discloses separating the image into separate files that contain only a portion of the

complete image … [, compressing] the partial image file … individually [, and]

transmitt[ing] [these partial files] separately to another location.” (EX1003 at 128-

35
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

129 (emphasis in original).) In contrast, the Applicant argued that the original

independent claims, including original claims 1 and 12, required “a single image file

that includes multiple compressed layers.” (EX1003 at 130.) The Applicant made

no arguments specific to the Examiner’s use of Lewis in view of Seroussi with regard

to the original claims 5 and 6. (EX1003 at 131.)

The Examiner and Applicant then went back and forth in several filings,

reiterating generally the same arguments regarding the teachings of Lewis.

Specifically, the Examiner responded in a second, final office action relying on the

same prior art and arguments as in the first office action. (EX1003 at 134-136.) The

Applicant then filed an after final response, again arguing that “the Lewis reference

does not disclose a complete image file that has compressed data.” (EX1003 at 141-

153, 142.) At this time, the Applicant further argued that Lewis does not use a

compression algorithm that is identified in a “key code” (as required by original

claims 1 and 12, see EX1003 at 71-72) because, instead, Lewis uses a “unique code

… to replace data” entirely. (EX1003 at 144; see also EX1003 at 144 (arguing that,

unlike the original claims, in Lewis, “a unique code is utilized to replace the data,

the code is transmitted and not the partial image data”).) The Examiner responded

in an advisory action, plainly stating that “Lewis and Seroussi disclose the claimed

invention, [where] specifically Fig. 2 of Lewis clearly shows a file including

compressed subregions and associated key codes.” (EX1003 at 155.)

36
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Following these filings largely repeating the same arguments, the Applicant

appealed. (EX1003 at 156.) In the appeal, the Applicant again argued that “the

Lewis reference does not disclose a complete image file that has compressed data

for different subregions” because the “Lewis reference discloses separating the

image into separate files or layers that contain only a portion of the complete file[.]”

(EX1003 at 167.) The Applicant emphasized these allegedly distinguishing features

of the original claims; for example, while original claim 1 required “compiling a

compressed data file including compressed data for each subregion and key code

identifying the compression algorithms applied to compress each subregion,” the

Applicant argued that Lewis instead uses separated data files (“layers”) where

“[t]hese two different layers are handled by completely different components and

are not a compressed data file.” (EX1003 at 169-170.)

The Examiner did not respond to this Appeal Brief and, instead, filed a new

non-final office action relying on new prior art. (EX1003 at 229.) The Examiner

relied on U.S. Patent No. 5,748,904 (“Huang,” EX1009) in place of Lewis, and again

relied on Seroussi as a secondary reference for original dependent claims 5 and 6.

The Applicant responded to the Examiner’s new non-office action by arguing—as

was similarly argued with respect to Lewis—that Huang “does not disclose a single

image file that includes multiple compressed subregions.” (EX1003 at 243.) The

Examiner then filed a final office action relying on the same art and rejections.

37
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

In response to the Examiner’s final office action, the Applicant for the first

time submitted proposed amendments to the original claims; the Applicant noted

these proposed amendments were to be discussed in “an Examiner Interview prior

to entry of the amendments.” (EX1003 at 258-264.) As shown below, the Applicant

proposed amending the original claim 1 to include original dependent claims 5 and

6, and original claim 12 to include original dependent claim 13. (EX1003 at 259-

261.) Prior to these proposed amendments, the interchange of arguments between

the Applicant and Examiner was substantively focused entirely on the independent

claims.

38
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(EX1001 at 259 (showing original dependent claims 5 and 6 being proposed to be

amended into original claim 1) (highlights added—the handwritten “5” and “6” are

original).)

(EX1001 at 260-261 (showing original dependent claim 13 being proposed to be

amended into original claim 12) (highlight added—the handwritten “13” is

original).)

An interview was held, and the Examiner filed an interview summary stating

that they will review the proposed amendments to determine allowability; the

Examiner provided no reasoning for potential allowability. (EX1003 at 266.)

Next, the Applicant filed an After Final Response in which amendments to

original claims 1 and 10 were entered. As shown below, additional amendments to

claim 1 as compared to those that were previously proposed were included (where

39
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

the differences between the original proposed amendments (shown above, see

EX1001 at 259) and what was added in these after final amendments is highlighted):

(EX1003 at 268 (annotation added added).)

For claim 10, however, the same amendment proposed before the interview

(shown above, see EX1001 at 260-261) was ultimately provided in the After Final

Response:

40
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(EX1003 at 269-270.)

The Applicant argued that the amendments in the After Final Response

rendered claim 1 allowable because Huang and Seroussi do not teach “selecting a

predictor algorithm from a plurality of predictor algorithms to generate predicted

pixel data values.” (EX1003 at 273-274.) The Applicant alleged that, instead,

Seroussi “only describes the use of a single predictor in the system,” and “with a

single predictor, a need does not exist to include any reference to the predictor

algorithm because only one predictor is utilized.” (EX1003 at 274.)

With regard to the original claim 12 (see EX1003 at 269-270 (where the

original claim 12 is renumbered as claim 10 for issuance)), the Applicant argued that

Huang does not disclose “a compression header including a cross reference of the

key code segments to the compression algorithm” and “appending a compression

header including a cross reference to the compression key codes” because Huang

41
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

only includes a “local header … associated with the each of the compressed

subregions” and “is devoid of a header for the image file” that includes a “cross

reference of the key code segments to the compression algorithms, as recited in

[original] claim 12.” (EX1003 at 275.)

In response, the Examiner responded with an advisory action simply stating

that the additional amendment to the original claim 1 (“selecting a predictor

algorithm from a plurality of predictor algorithms to generate predicted pixel data

values”) “raises a new issue, which would require further consideration and search.”

(EX1003 at 279-280.)

The Applicant then filed a Request for Continued Examination based on the

amendments listed above. Following the RCE’s filing, the Examiner allowed the

claims. (EX1003 at 282-285.) The Examiner provided the following reasons for

allowance:

42
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(EX1003 at 284.)

Thus, for Claim 1, the ʼ583 Patent’s prosecution history suggests that the

Examiner did not identify prior art teaching “selecting a predictor algorithm from a

43
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

plurality of predictor algorithms” and a “compression header including predictor

data identifying the selected predictor algorithm.” (EX1003 at 274, 284; EX1001 at

Claim 1.) And for Claim 10, the ʼ583 Patent’s prosecution history suggests that the

Examiner did not identify prior art teaching “a compression header including a cross

reference of the key code segments to the compression algorithms.” (EX1003 at

275, 284; EX1001 at Claim 10.) As demonstrated below, however, these aspects

were known in the art.

E. Level of Ordinary Skill in the Art

A person of ordinary skill in the art (“POSA”) at the time of filing of the ’583

patent (Nov. 24, 1999) would have had at least the equivalent of a Bachelor’s degree

in electrical engineering, computer engineering, computer science, or a related

subject and two or more years of experience in the field of image/video coding. Less

work experience may be compensated by a higher level of education, such as a

Master’s Degree, and vice versa. (EX1002 at ¶¶75-81.)

V. CLAIM CONSTRUCTION
The ’583 Patent is expired. (EX1001 at Cover.) For the reexamination of

expired patents, the MPEP requires:

In a reexamination proceeding involving claims of an expired patent, claim

construction pursuant to the principle set forth by the court in Phillips v. AWH
Corp., 415 F.3d 1303, 1316, 75 USPQ2d 1321, 1329 (Fed. Cir. 2005) (words
of a claim ‘are generally given their ordinary and customary meaning’ as

44
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

understood by a person of ordinary skill in the art in question at the time of

the invention) should be applied since the expired claim are not subject to
amendment.
(MPEP § 2258 (citing Ex parte Papst-Motoren, 1 USPQ2d 1655 (Bd. Pat. App. &

Inter. 1986).)

Thus, the ʼ583 Patent’s Claims should be given their ordinary and customary

meaning as understood by a person of ordinary skill in the art in question at the time

of the invention.

VI. SUBSTANTIAL NEW QUESTION OF PATENTABILITY #1 – H.263+

RENDERS CLAIMS 1 AND 10 OBVIOUS

A. Overview of Prior Art Printed Publications Raising Substantial

New Questions of Patentability

1. H.263+ (Ex. 1004)

Originally ratified in March 1996 by the International Telecommunication

Union’s Telecommunication Standardization Sector (ITU-T), the H.263 standard is

a well-known video coding standard. (EX1002, ¶83; see supra Section III.)

Following its original ratification in March 1996, the ITU-T published revisions

providing various improvements and extensions. (Id.) H.263 version 2 (provided

as the H.263+ prior art reference herein (EX1004) and commonly referred to as

H.263+ or H.263v2) was published in February 1998 and included extensions in

Annexes I–T. (EX1002, ¶83; see generally H.263+.)

45
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

H.263+ in Annex I provides an “Advanced Intra Coding mode.” (H.263+ at

73-80.) And as was described in Section III above, H.263+, and particularly the

Advanced Intra Coding mode of Annex I, relies on selected prediction and

compression algorithms that are identified in the syntax of coded image data in

picture and macroblock headers.

Additionally, the Examiner did not consider the H.263 standard or H.263+

during prosecution of the ʼ583 Patent. (See generally EX1003.)

i. H.263+ Is a Prior Art Printed Publication

H.263+ qualifies as prior art under at least 35 U.S.C. §§ 102(a) and 102(b).

(EX1002 at ¶¶84-89.)

As explained by the Federal Circuit, “public accessibility has been called the

touchstone in determining whether a reference constitutes a printed publication.” In

re Hall, 781 F.2d 897, 899 (Fed. Cir, 1986). “A reference is considered publicly

accessible if it was disseminated or otherwise made available to the extent that

persons interested and ordinarily skilled in the subject matter or art, exercising

reasonable diligence, can locate it.” Acceleration Bay, LLC v. Activision Blizzard,

Inc., 908 F.3d 765, 772 (Fed. Cir. 2018).

Petitioner submits the Declaration of Dr. Havlicek, who explains that H.263+

was publicly accessible at least one year before the filing date of the ʼ583 Patent.

(EX1002, ¶¶84-89.)

46
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Specifically, as Dr. Havlicek explains, H.263+ was available online on the

website of the International Telecommunication Union—Telecommunication

Standardization Sector (ITU-T) at least by October 22, 1998. (EX1002, ¶¶84-89.)

Dr. Havlicek further explains that a person of ordinary skill in the art (“POSA”) in

the field of image and video compression would have been aware of the ITU-T

website and would have frequently accessed, without restriction, developments on

the ITU-T website pertaining to the H.263 standard as the industry standard for

image and video compression technology. (EX1002, ¶87.) Dr. Havlicek explains

that the ITU-T, its website, and the development of the H.263 standard were well-

known to those interested in image and video compression. (EX1002, ¶87; Voter

Verified, Inc. v. Premier Election Sols., 698 F.3d 1374, 1381 (Fed. Cir. 2012)

(testimony indicating that an online publication or website on which a reference was

published was well-known and accessible to a POSA can support a finding of public

accessibility).)

For example, the prior art includes many citations to the H.263+ standard.

(EX1002, ¶88.) For example, “H.263+: Video Coding at Low Bit Rates” (“Cote”

EX1003) by Guy Cote et al. and published by the IEEE in Nov. 1998 is prior art

describing the H.263+ standard.

47
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Further, Dr. Havlicek explains that international standards such as H.263 are

compilations of the contributions of POSAs throughout the world and are relied on

by POSAs. (EX1002, ¶87 (citing EX1012, 1; EX1013, 1; EX1014, 1).)

Thus, H.263+ was publicly available more than one year prior to November

24, 1999, the ʼ583 Patent’s filing date, and, thus, H.263+ qualifies as prior art under

at least 35 U.S.C. §§ 102(a) and 102(b). (EX1002, ¶¶84-89; EX1001, Cover Page.)

2. Supporting Reference: Cote (Ex. 1005)

“H.263+: Video Coding at Low Bit Rates” is an article by Guy Cote et al. that

was published by the IEEE in November 1998. (EX1005, “Cote”.) Cote was not

considered during prosecution of the ʼ583 Patent. (See generally EX1003.) Cote

provides an overview of the H.263+ standard (where the H.263+ reference includes

the published standard) and additional insight into its requirements. (EX1002 at ¶90;

see generally EX1005.) Because of this, Cote is cited in the Rejections below to

further support and explain the operations of H.263+.

Cote qualifies as prior under at least 35 U.S.C. § 102(a). (EX1002 at ¶¶91-

94.) As explained with regard to the H.263+ reference above, a reference qualifies

as prior art if it was publicly accessible prior to the ʼ583 Patent’s filing. See In re

Hall, 781 F.2d at 899; Acceleration Bay, 908 F.3d at 772.

Petitioner submits the Declaration of Dr. Havlicek, who explains that Cote was

publicly accessible prior to the filing data of the ʼ583 Patent. (EX1002 at ¶91-94.)

48
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Specifically, Dr. Havlicek explains that Cote was published in the IEEE

Transactions on Circuits and Systems for Video Technology, Volume 8, No. 7 in

November 1998. (EX1002 at ¶91; EX1005 at 849.) Persons interested and

ordinarily skilled in the subject matter or art, exercising reasonable diligence, could

locate an article published by the IEEE, as the IEEE is a preeminent source for

publicly accessible articles in the art of image/video coding. (EX1002 at ¶93; see

also Acceleration Bay, 908 F.3d at 772.) Dr. Havlicek explains that the IEEE and

the publications of the IEEE journals were well-known to those interested in image

and video compression. (EX1002, ¶¶941-94; Voter Verified, Inc. v. Premier

Election Sols., 698 F.3d 1374, 1381 (Fed. Cir. 2012) (testimony indicating that an

online publication or website on which a reference was published was well-known

and accessible to a POSA can support a finding of public accessibility).)

Thus, Cote was publicly available as of November 1998, which is prior to the

ʼ583 Patent’s filing date of November 24, 1999, and, thus, Cote qualifies as prior art

at least under 35 U.S.C. § 102(a). (EX1002 at ¶¶91-94.)

B. H.263+ Renders Claims 1 and 10 Obvious

The Examiner did not consider H.263+ or Cote during prosecution of the ʼ583

Patent. (See generally EX1003.) The Examiner similarly did not consider the H.263

Standard’s original 1996 publication during prosecution. (Id.) Moreover, Lewis,

Seroussi, and Huang—the three patents relied on in rejections during prosecution—

49
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

do not even mention the H.263 standard. (See generally EX1007-1009.) And as

shown below, H.263+ renders obvious each of the limitations of Claims 1 and 10 of

the ʼ583 Patent and, thereby, raises substantial new questions of patentability.

Particularly, while the Examiner allowed Claim 1 following an amendment

that required a selected predictor algorithm being identified in a compression header

(EX1003 at 274, 284), H.263+ plainly requires such.

Additionally, while the Examiner allowed Claim 10 following an amendment

that required that a compressed image file include a compression header including a

cross reference of the key code segments to the compression algorithm (EX1003 at

275, 284), H.263+ plainly requires such.

1. Proposed Rejection: H.263+ renders Claim 1 obvious

“Obviousness is a question of law based on underlying factual findings

relating to ‘the scope and content of the prior art, differences between the prior art

and the claims at issue, the level of ordinary skill in the pertinent art, and any

objective indicia of non-obviousness.’” Paice LLC v. Ford Motor Co., 881 F.3d

894, 900 (Fed. Cir. 2018) (quoting Randall Mfg. v. Rea, 733 F.3d 1355, 1362 (Fed.

Cir. 2013)). A claim is obvious when all of its limitations are disclosed in the prior

art, there is a reason that would have prompted a person of ordinary skill in the art

to combine or modify the prior art so as to achieve the claimed invention, and there

is a reasonable expectation of success. KSR Int’l Co. v. Teleflex, Inc., 550 U.S. 398,

50
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

418 (2007); Genzyme Therapeutic Prods. Ltd. v. Biomarin Pharm. Inc., 825 F.3d

1360, 1373 (Fed. Cir. 2016).

As shown below, and as further explained by Dr. Joseph Havlicek, an expert

experienced in the field of image and video compression (Havlicek Decl. ¶¶95-133),

H.263+ renders Claim 1 obvious.

i. A method for compressing an image data stream, the

method comprising the steps of:
Claim 1’s preamble provides a broader statement of purpose (compressing an

image data stream) that is further defined by the Claim’s body. Thus, the preamble

should not be understood to require additional limitations. This point is most evident

when considering the transitional phrase employed between the preamble and the

body; the preamble generally presents “a method for compressing an image data

stream” before transitionally stating “the method comprising the steps of: [body of

the claim].” Thus, the broader purpose defined in the Claim’s preamble is further

described (i.e., “the method comprising the steps of…”) by the Claim’s body.

Even if the preamble is limiting, H.263+ discloses or at least renders obvious

a method for compressing an image data stream. H.263+ describes a video coding

standard, where video data is made up of an image data stream. (H.263+ at 7 (“The

[H.263] source code operates on non-interlaced pictures….”)). H.263+ refers to the

51
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

individual images of an image data stream as “pictures.” (Id.; see also EX1002,

¶97.)

Additionally, H.263+ explains that the purpose of the H.263 standard is

“improved compression performance” for video coding. (H.263+ at Cover

Summary (PDF page 3).) For example, H.263+ explains that the Advanced Intra

Coding mode of Annex I “significantly improves the compression performance over

the INTRA coding of the core H.263 syntax.” (H.263+ at 4; see also EX1002, ¶98.)

Thus, a person of ordinary skill in the art (POSA) would have understood

H.263+ as describing a method for compressing an image data stream (EX1002,

¶¶96-98.)

ii. (a) dividing the image data stream into a plurality of

subregions,
H.263+ discloses or at least renders obvious this limitation.

As illustrated in Figure 2 of Cote below, H.263+ describes dividing pictures

into Groups of Blocks (GOBs), macroblocks, and blocks. (H.263+ at 9-11, 9 (“Each

picture is divided either into Groups of Blocks (GOBs) or into slices.”), 11 (“Each

GOB is divided into macroblocks. … Further, a macroblock consists of four

luminance blocks and the two spatially corresponding colour difference blocks as

shown in Figure 5.”); see also H.263+ at Figure 5/H.263 (page 11) (showing the

“Arrangement of blocks in a macroblock”); EX1002 at ¶99; Cote at 849 (“Each

52
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

picture in the input video sequence is divided into macroblocks, consisting of four

luminance blocks of 8 pixels x 8 lines followed by one Cb block and one Cr block,

each consisting of 8 pixels x 8 lines. A group of blocks (GOB) is defined as an

integer number of macroblock rows, a number that is dependent on picture

resolution. For example, a GOB consists of a single macroblock row at QCIF

resolution.”).) Thus, a POSA would have understood the macroblocks of H.263+ as

corresponding to subregions of the picture/image. (EX1002 at ¶99.)

53
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(Cote at 849 (Figure 2))

Thus, a POSA would have understood H.263+ as teaching “dividing the image

data stream into a plurality of subregions” as is claimed, where the claimed

“subregions” correspond to macroblocks. (EX1002 at ¶¶99-100.)

54
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

iii. (b) selecting a predictor algorithm from a plurality of

predictor algorithms to generate predicted pixel data
values …
H.263+ discloses or at least renders obvious this limitation.

H.263+ predicts images according to a prediction algorithm selected from a

plurality of predictor algorithms. Specifically, H.263+ explains that images can be

predicted according to different prediction types, including P-picture (inter) and I-

picture (intra) prediction. (H.263+ at 11 (“INTER: A picture using a temporally

previous reference picture (also called a P-picture); … INTRA: A picture having no

reference picture(s) for prediction (also called an I-picture)”).) P-pictures (inter) rely

on “temporally previous” pictures to make predictions. (H.263+ at 11; Cote at 851-

852, 854-855, Figure 7a (describing inter-predicted pictures including P-type

pictures, which rely on a single temporally previous picture).) I-pictures (intra) do

not rely on other pictures for prediction. (H.263+ at 11 (“[t]he [prediction] coding

mode in which temporal prediction is applied is called INTER; the coding mode is

called INTRA if no temporal prediction is applied”).) Instead, I-pictures may rely

on macroblocks within the same picture for prediction. For example, when the

“Advanced Intra Coding mode” is in use, I-pictures include intra macroblocks that

are predicted based on macroblocks within the same picture. (H.263+ at 4 (in the

Advanced Intra Coding mode, “intra blocks are first predicted from neighboring

intra blocks [within the same picture] prior to coding”); EX1002 at ¶102.)

55
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

In addition to H.263+’s description that predictor algorithms can be defined

in the picture layer, H.263+ also describes defining predictor algorithms in the

macroblock layer. (EX1002 at ¶103.) For example, as seen in Table 9 below,

H.263+ provides that I-pictures (“INTRA” picture type, as seen in the left column)

and P-pictures (“INTER” picture type) can include different types of macroblocks,

numbered 1-5, for which different inter and intra prediction algorithms are defined.

(H.263+ at Table 9 (Page 36).)

I-pictures (“Picture type[:] INTRA”) can include only intra macroblocks (MB types

3-4). (H.263+ at Table 9 (Page 36) (providing that INTRA pictures (I-pictures) can

rely on “MB type[s]” 3-4, where these “MB type[s]” are called “INTRA,” and

“INTRA+Q”).) P-pictures (“Picture type[:] INTER”) can include both intra and

inter macroblocks (MB types 0-5). (H.263+ at Table 9 (Page 36) (providing that

56
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

INTER pictures (P-pictures) can rely on “MB type[s]” 0-5, where these “MB

type[s]” are called “INTER,” “INTER+Q,” “INTER4V,” “INTRA,” “INTRA+Q,”

and “INTER4V+Q”); EX1002 at ¶104.)

Thus, while P-pictures are “INTER” pictures, (H.263+ at 11), P-pictures can

include both inter and intra macroblocks. (H.263+ at 11-12 (“The coding mode in

which temporal prediction is applied is called INTER; the coding mode is called

INTRA if no temporal prediction is applied. The INTRA coding mode can be

signalled [sic] at the picture level (INTRA for I-pictures or INTER for P-pictures)

or at the macroblock level in P-pictures.”; EX1002 at ¶105; Cote at 852 (“The two

switches in Fig. 1 represent the intra/inter mode selection, which is not specified in

the standard. Such a selection is made at the macroblock level.”).)

H.263+ further describes selected predictor algorithms being identified in the

syntax of the picture and macroblock layers in headers. In the case of the Advanced

Intra Coding mode (see supra Section III.A.1 (describing the Advanced Intra

Coding mode)) and as illustrated below, H.263+ describes a compressed image file

including a picture layer (comprising a picture header and GOBs), a GOB layer

(comprising a GOB header and macroblock data), and a macroblock layer

(comprising a macroblock header and block data). (H.263+ at Figures 7 (Page 22),

8 (Page 22) (annotated to include MPPTYPE and OPPTYPE subfields), 9 (Page 32),

and I.1 (Page 74); see also EX1002 at ¶106.)

57
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at Figures 7 (Page 22), 8 (Page 22) (annotated to include MPPTYPE and

OPPTYPE subfields), 9 (Page 32), and I.1 (Page 74).)

H.263+ describes the picture header and macroblock header as including

information which identifies selected predictor algorithms. Specifically, as

illustrated above, H.263+ explains that, when the Advanced Intra Coding mode is

applied, the picture header includes a “PLUSPTYPE” element including MPPTYPE

and OPPTYPE subfields. (H.263+ at 11-12, 22-25; EX1002 at ¶107.) The

58
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

MPPTYPE subfield of the PLUSPTYPE syntax element identifies a selected

“picture coding type”—including whether the picture is an “INTRA (I-picture)” or

“INTER (P-picture).” (H.263+ at 25 (providing that the MPPTYPE subfield of the

PLUSPTYPE element identifies “Picture Type Code” in “Bits 1-3”); EX1002 at

¶107; see also supra Section III.A.2 (explaining that “when the PLUSPTYPE

element is present, the PLUSPTYPE element identifies picture type instead of the

PTYPE element”).)

Additionally, the encoder selects the picture type (as identified in the

MPPTYPE subfield of the PLUSPTYPE syntax element) on a picture by picture

basis. (H.263+ at 25 (“The mandatory part of PLUSPTYPE [which includes the

MPPTYPE subfield] consists of features which are likely to be changed on a

picture-by-picture basis. Primarily, these are the bits to indicate the picture type

among I, P, [and other picture types.]”); EX1002 ¶108.)

Thus, a POSA would have understood H.263+ as describing selectable

predictor algorithms, those selectable predictor algorithms comprising P-pictures

(inter) and I-pictures (intra), (see H.263+ at 11), where these predictor algorithms

are identified in the picture header in the MPPTYPE subfield of the PLUSPTYPE

element (see H.263+ at 22-25). (EX1002 at ¶¶101-109.)

And, thus, while the Applicant argued during prosecution that Seroussi

(EX1008) “only describes the use of a single predictor in the system” (see EX1003

59
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

at 274), it would have been clear to a POSA that H.263+ describes the use of

multiple predictors—including P-pictures (inter) and I-pictures (intra)—in the

coding system. (EX1002 at ¶110.)

H.263+ further provides that the predictor algorithms generate predicted pixel

data values. (EX1002 at ¶111; H.263+ at 11 (explaining that each macroblock

relates to blocks each comprising a set number of pixels), 45 (describing “the pixels

being predicted”).)

Thus, a POSA would have understood H.263+ as describing “selecting a

predictor algorithm from a plurality of predictor algorithms to generate predicted

pixel data values.” (EX1002 at ¶¶101-112.)

iv. and compressing data of each subregion in accordance

with one of a plurality of compression algorithms;
H.263+ discloses or at least renders obvious this limitation.

When the Advanced Intra Coding mode is not applied, H.263+ compresses

all macroblocks (subregions) using the same VLC table. (EX1002 at ¶114; see Cote

at 851 (describing the use of entropy coding in H.263), 854 (when the Advanced

Intra Coding mode is not applied, explaining that “the standard employs the same

VLC table for coding all quantized coefficients”); H.263+ at 11 (“the [predicted]

coefficients are quantized and entropy coded”), 41-44 (including in Table 16 the

VLC table for coefficients)).

60
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

When the Advanced Intra Coding mode is applied, H.263+ compresses

macroblocks based on two VLC tables—a first VLC table that is applied to inter

macroblocks, and a second VLC table that is applied to intra macroblocks. (EX1002

at ¶115; H.263+ at 4 (“Separate Variable Length Code (VLC) tables are defined for

the INTRA blocks”), 74-75 (using a “separate VLC for INTRA coefficients”).)

H.263+’s reliance on a different compression algorithm for intra macroblocks in the

Advanced Intra Coding mode optimizes the compression of intra macroblocks.

(EX1002 at ¶115; Cote at 854 (“The main part of the [H.263+] standard employs the

same VLC table for coding all quantized coefficients. However, this table is

designed for inter macroblocks and is not very effective for coding intra

macroblocks.”), id. at 854 (explaining that using different VLC tables for the intra

macroblocks in the Advanced Intra Coding mode “is optimized to global statistics

of intra macroblocks” where “larger coefficients with smaller runs of zeros are more

common”).)

As illustrated below, H.263+ indicates whether the Advanced Intra Coding

mode is in use based on the OPPTYPE subfield of the PLUSPTYPE element in the

picture header. (H.263+ at 24-25 (in describing the syntax of the PLUSPTYPE field,

explaining that a “subfield” termed “OPPTYPE” can be present that identifies

whether the “Advanced Intra Coding mode (AIC) mode” is being used), 73 (“The

61
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

use of [the Advanced Intra Coding mode] is indicated in the PLUSPTYPE field of

the picture header.”); see also EX1002 at ¶116.)

(H.263+ at Figures 7 (Page 22), 8 (Page 22) (annotated to include

MPPTYPE and OPPTYPE subfields), 9 (Page 32), and I.1 (Page 74).)

Additionally, as illustrated above, H.263+ provides that the macroblock type

(inter or intra) is identified in the macroblock header’s MCBPC syntax element.

62
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at 33-36 (providing that the macroblock header includes a MCBPC syntax

element which defines the “macroblock type” including whether the macroblock is

an “INTRA” or “INTER” type macroblock; see supra Section VI.B.1.iii (describing

the different types of INTRA and INTER macroblocks in the discussion of H.263+’s

Table 9); see also EX1002 at ¶117)

Thus, a POSA would have understood that when the Advanced Intra Coding

mode is applied—as is identified in the OPPTYPE subfield of the PLUSPTYPE

field of the picture header—H.263+ selects between two VLC tables to compress

macroblocks based on macroblock type as is identified in the MCBPC syntax

element of the macroblock header. (EX1002 at ¶¶113-118.) A POSA would have

further understood that when the Advanced Intra Coding mode is applied, H.263+

selects a first VLC table for intra macroblocks, and a second VLC table for inter

macroblocks. (EX1002 at ¶¶113-118.)

Moreover, a POSA would have understood that using different VLC tables

for different macroblock-types (intra and inter) amounts to using different

compression algorithms for different macroblock-types—one compression

algorithm for each VLC table. (EX1002 at ¶119; see also Cote at 851 (“Entropy

coding [(a type of compression encoding)] is performed by means of variable-

length-codes (VLC’s).”), 854 (“Thus, the advanced intra coding mode employs a

63
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

new VLC table for encoding [(compressing)] the quantized coefficients, a table that

is optimized to global statistics of intra macroblocks.”).)

Thus, a POSA would have understood H.263+ as describing “compressing

data of each subregion in accordance with one of a plurality of compression

algorithms.” (EX1002 at ¶¶113-120.)

v. (c) compiling a compressed data file including

compressed data for each subregion and key code
identifying the compression algorithms applied to
compress each subregion, …
H.263+ discloses or at least renders obvious this limitation.

As seen below, H.263+ describes compiling a compressed picture, which is

inclusive of all of the data of the compressed macroblocks (subregions). (H.263+ at

49 (with regard to buffering data for transmission/receiving, explaining that a

“complete coded picture is in the buffer” where the corresponding buffers are seen

in Figure 1 on page 2 of H.263+); see also EX1002 at ¶121.)

64
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at Figures 7 (Page 22), 8 (Page 22) (annotated to include

MPPTYPE and OPPTYPE subfields), 9 (Page 32), and I.1 (Page 74).)

Thus, while the Applicant argued repeatedly during prosecution that “the

Lewis reference does not disclose a complete image file that has compressed data”

(EX1003 at 141-153, 142; see also EX1003 at 167, 169-170), it would have been

clear to a POSA that H.263+ describes a completely compressed image file

including the compressed data for each subregion. (EX1002 at ¶122.)

65
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Additionally, as is also explained above in Section VI.B.1.iv and as illustrated

above, H.263+ provides that the macroblock type (inter or intra) is identified in the

macroblock header’s MCBPC syntax element. (H.263+ at 33-36 (providing that the

macroblock header includes a MCBPC syntax element which defines the

“macroblock type” including whether the macroblock is an “INTRA” or “INTER”

type macroblock). And as is also explained above in Section VI.B.1.iv, when the

Advanced Intra Coding mode is applied—as is identified in the OPPTYPE subfield

of the PLUSPTYPE element of the picture header (see H.263+ at 23-25, 73)—the

compression algorithm (VLC table) is selected based on whether a macroblock is

intra or inter type as is identified in the MCBPC syntax element of the macroblock

header. (EX1002 at ¶123; H.263+ at 4 (“Separate Variable Length Code (VLC)

tables are defined for the INTRA blocks”), 74-75 (using a “separate VLC for INTRA

coefficients”).)

Moreover, a POSA would have understood that the MCBPC syntax element

corresponds to a “key code identifying the compression algorithms applied to

compress each subregion” (as is claimed) because, as explained in Section VI.B.1.iv

above, the MCBPC syntax element identifies the VLC table (compression

algorithm) applied to compress each macroblock (subregion) when the Advanced

Intra Coding mode is applied. (See EX1002 at ¶¶113-120, 124.)

66
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Thus, a POSA would have understood that H.263+ describes “compiling a

compressed data file including compressed data for each subregion and key code

identifying the compression algorithms applied to compress each subregion.”

(EX1002 at ¶¶121-125.)

vi. wherein the compressed data comprises pixel data that

includes predictor error data representative of
differences between predicted pixel data values and
actual pixel data values, and
H.263+ discloses or at least renders obvious this limitation.

H.263+ provides that predictions are representative of differences between

predicted blocks and originally received blocks (the “actual” blocks). (H.263+ at 9

(in Figure 3, which describes the H.263 Source Coder, including a subtractor block

applied during prediction), 73-74 (“A particular INTRA-code block may be

predicted from the block above the current block being decoded”); Cote at 849 (with

regard to H.263, explaining that “only the prediction error frames—the difference

between original frames and motion-compensated predicted frames—need be

encoded”).) Particularly, H.263+ relies on DCT coefficients to make predictions,

where the DCT coefficients are representative of the pixels of the original block.

(Cote at 854 (with regard to the Advanced Intra Coding mode, describing “[t]he

difference coefficients, obtained by subtracting the predicted DCT coefficients from

the original ones” where the “coefficients” are DCT coefficients representative of

67
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

the pixels in an original block (see Cote at 851 (“the 8 x 8 DCT specified in H.263

is to decorrelate the 8 x 8 blocks of original pixels”))); EX1002 at ¶126.) H.263+

further describes that the original blocks (prior to decorrelation to generate the DCT

coefficients) are made up of pixels. (H.263+ at 11 (“Each … block thus relates to 8

pixels by 8 lines….”); EX1002 at ¶126.)

Thus, a POSA would have understood that predictor algorithms in H.263+

involve differences being calculated (“predictor error data”) that is representative of

the differences between predicted pixel data values (from a predicted block) and the

actual pixel data values (from the original/actual block). (EX1002 at ¶127.) Indeed,

the DCT coefficients are representative of the pixels in an original block. (EX1002

at ¶126.)

Thus, a POSA would have understood that H.263+ teaches “wherein the

compressed data comprises pixel data that includes predictor error data

representative of differences between predicted pixel data values and actual pixel

data values.” (EX1002 at ¶¶126-128.)

vii. (d) appending a compression header to the compressed

data file, the compression header including predictor
data identifying the selected predictor algorithm used
to generate the predicted pixel data values.
H.263+ discloses or at least renders obvious this limitation.

68
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

As is illustrated below, H.263+ provides that a picture header is included with

(and thereby, during the encoding process, is appended to) a compressed picture file.

(H.263+ at Figures 7 (Page 22), 8 (Page 22) (annotated to include MPPTYPE and

OPPTYPE subfields), 9 (Page 32), and I.1 (Page 74).) Indeed, H.263+ describes

buffering data during encoding, where buffering comprises appending data to a file

until it is ready for transmission. (EX1002 at ¶129; see H.263+ at Figure 1 (page 2)

(showing the use of a “transmission buffer” and “receiving buffer”), 6 (describing

the “buffering” process for each picture), 49 (with regard to buffering data for

transmission/receiving, explaining that a “complete coded picture is in the buffer”

where the corresponding buffers are seen in Figure 1 on page 2 of H.263+).) Thus,

a POSA would have understood that, because H.263+ describes a compressed

picture that includes a picture header and compressed data, the picture header must

be appended to the compressed data in forming the compressed picture (as depicted

in H.263+, Figure 7) prior to transmission. (EX1002 at ¶129.) Moreover,

“appending” does not place additional requirements, such as temporal requirements,

on how the data is combined—appending only requires that the data (the

“compression header” and compressed data file) are combined in the compressed

image data stream.

And as was explained in Section VI.B.1.iii above, when the Advanced Intra

Coding mode is applied, the picture header includes the MPPTYPE subfield of the

69
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

PLUSPTYPE syntax element, which identifies a selected predictor algorithm

(“picture coding type”)—including whether the picture is an “INTRA (I-picture)” or

“INTER (P-picture).” (H.263+ at 25 (providing that the MPPTYPE subfield of the

PLUSPTYPE element identifies “Picture Type Code” in “Bits 1-3” including “I-

picture (INTRA)” and “P-picture (INTER)”); EX1002 at ¶130; see also supra

Section III.A.2 (explaining that “when the PLUSPTYPE element is present, the

PLUSPTYPE element identifies picture type instead of the PTYPE element”).)

70
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at Figures 7 (Page 22), 8 (Page 22) (annotated to include

MPPTYPE and OPPTYPE subfields), 9 (Page 32), and I.1 (Page 74).)

A POSA would have understood the picture header as corresponding to a

“compression header including predictor data identifying the selected predictor

algorithm used to generate the predicted pixel values” (as is claimed) because the

picture header includes predictor data (PLUSPTYPE and MPPTYPE) that identifies

the selected predictor algorithm (i.e., I-picture or P-picture) used to generate the

predicted pixel values. (EX1002 at ¶131.) Additionally, the picture header

(“compression header”) includes information related to the compression algorithm

applied because, as explained in Section VI.B.1.iv, the picture header includes the

OPPTYPE syntax element identifying whether the Advanced Intra Coding mode is

applied; where, as explained in Section VI.B.1.v, when the Advanced Intra Coding

mode is applied, compression algorithm selection (i.e., VLC table selection) occurs

based on macroblock type. (EX1002 at ¶131.)

Thus, a POSA would have understood H.263+ as appending a picture header

(“compression header”) to the compressed data file (where the “compressed data

file” is described above in Section VI.B.1.v) where the picture header includes the

MPPTYPE subfield of the PLUSPTYPE element identifying a selected predictor

algorithm (i.e., I-picture, P-picture) used to generated predicted pixel data values.

(EX1002 at ¶132.)

71
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

And, thus, a POSA would have understood H.263+ as teaching “appending a

compression header to the compressed data file, the compression header including

predictor data identifying the selected predictor algorithm used to generate the

predicted pixel data values.” (EX1002 at ¶¶129-133.)

2. Proposed Rejection: H.263+ renders Claim 10 obvious

As shown below, and as further explained by Dr. Joseph Havlicek, an expert

experienced in the field of image and video compression (EX1002 at ¶¶134-167),

H.263+ renders Claim 10 obvious.

i. A compressed image data file comprising:

Claim 10’s preamble provides a broader statement of purpose (a compressed

image data file) that is further defined by the Claim’s body. Thus, the preamble

should not be understood to require additional limitations. This point is most evident

when considering the transitional phrase employed between the preamble and the

body; the preamble generally presents “a compressed image data file” before

transitionally stating “comprising: [body of the claim].” Thus, the broader purpose

defined in the Claim’s preamble is further described (i.e., “comprising: …”) by the

Claim’s body.

Even if the preamble is limiting, H.263+ discloses or at least renders obvious

a compressed image data file. H.263+ describes a video coding standard, where the

coded video data is made up of compressed images. (H.263+ at 7 (“The [H.263]

72
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

source code operates on non-interlaced pictures….”)). H.263+ refers to the

individual images of an image data stream as “pictures.” (Id.) For example, as seen

below, H.263+ describes compiling a compressed picture which is inclusive of all

of the data of the compressed macroblocks (subregions). (H.263+ at 49 (with regard

to buffering data for transmission/receiving, explaining that a “complete coded

picture is in the buffer” where the corresponding buffers are seen in Figure 1 on

page 2 of H.263+); EX1002 at ¶135)

73
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at Figures 7 (Page 22), 8 (Page 22) (annotated to include

MPPTYPE and OPPTYPE subfields), 9 (Page 32), and I.1 (Page 74).)

Moreover, H.263+ describes that the compressed image being buffered in its

entirety (and, thereby, a compressed image (i.e., picture) file). (H.263+ at 49 (with

regard to buffering data for transmission/receiving, explaining that a “complete

coded picture is in the buffer” where the corresponding buffers are seen in Figure 1

on page 2 of H.263+); EX1002 at ¶136)

Thus, a POSA would have understood H.263+ as describing a compressed

image data file. (EX1002, ¶¶135-137.)

ii. a plurality of compressed data segments, each segment

including compressed data code for a subregion of an
image data stream;
H.263+ discloses or renders obvious this limitation.

As illustrated in Figure 2 of Cote below, H.263+ describes dividing pictures

into Groups of Blocks (GOBs), macroblocks, and blocks. (H.263+ at 9-11, 9 (“Each

picture is divided either into Groups of Blocks (GOBs) or into slices.”), 11 (“Each

GOB is divided into macroblocks. … Further, a macroblock consists of four

luminance blocks and the two spatially corresponding colour difference blocks as

shown in Figure 5.”); see also H.263+ at Figure 5/H.263 (page 11) (showing the

“Arrangement of blocks in a macroblock”); EX1002 at ¶138; Cote at 849 (“Each

picture in the input video sequence is divided into macroblocks, consisting of four

74
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

luminance blocks of 8 pixels x 8 lines followed by one Cb block and one Cr block,

each consisting of 8 pixels x 8 lines. A group of blocks (GOB) is defined as an

integer number of macroblock rows, a number that is dependent on picture

resolution. For example, a GOB consists of a single macroblock row at QCIF

resolution.”).) Thus, a POSA would have understood the macroblocks of H.263+ as

corresponding to subregions of an image data stream (a stream of pictures

comprising video). (EX1002 at ¶138.)

75
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(Cote at 849 (Figure 2))

Additionally, referring to the compiled figures of H.263+ below, H.263+

describes the syntax of the compressed data stream as including a picture layer

(comprising a picture header and GOBs), a GOB layer (comprising a GOB header

and macroblock data), and a macroblock layer (comprising a macroblock header and

block data). (H.263+ at Figures 7 (Page 22), 9 (Page 32), and I.1 (Page 74); EX1002

at ¶139.)

76
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at Figures 7 (Page 22), 8 (Page 22) (annotated to include

MPPTYPE and OPPTYPE subfields), 9 (Page 32), and I.1 (Page 74).)

Referring to the figure above, the macroblocks—where multiple macroblocks

are included in a picture (see, e.g., H.263+ at 11 (“Each GOB is divided into

macroblocks”))—each include “block data” that is compressed based on the coding

algorithms of H.263+. (EX1002 at ¶140; see Cote at 857 (explaining that H.263

77
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Advanced Intra Coding mode “significantly improves compression of intra

macroblocks”).) Thus, a POSA would have understood each of the subregions of

H.263+ (macroblocks) as including compressed data segments (the compressed

block data of the macroblock). (EX1002 at ¶140.)

Thus, a POSA would have understood H.263+ as describing “a plurality of

compressed data segments, each segment including compressed data code for a

subregion of an image data stream.” (EX1002 at ¶¶138-141.)

iii. a plurality of compression key code segments, each key

code segment being associated with at least one
compressed data segment …
H.263+ discloses or renders obvious this limitation.

As seen below, in Table 9, H.263+ describes six different types (“MB Type”

numbered 0-5) of macroblocks including different types of inter and intra

macroblocks (as indicated under the “Name” column). (H.263+ at 33-36; Cote at

852 (“The two switches in Fig. 1 represent the intra/inter mode selection, which is

not specified in the standard. Such a selection is made at the macroblock level.”);

EX1002 at ¶143.)

78
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at Table 9 (Page 36).)

Macroblock headers indicate macroblock type. (EX1002 at ¶144.) As

illustrated below, H.263+ provides that the macroblock type (inter or intra) is

identified in the macroblock header’s “MCBPC” syntax element. (H.263+ at 33-36

(providing that the macroblock header includes a “MCBPC” syntax element which

defines the “macroblock type” including whether the macroblock is an “INTRA” or

“INTER” type macroblock).

79
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at Figures 7 (Page 22), 8 (Page 22) (annotated to include

MPPTYPE and OPPTYPE subfields), 9 (Page 32), and I.1 (Page 74).)

As is further illustrated above, for the Advanced Intra Coding mode (see supra

Section III.A.1 (describing the Advanced Intra Coding mode)), H.263+ indicates

whether the Advanced Intra Coding mode is being applied based on the OPPTYPE

subfield of the PLUSPTYPE element in the picture header. (H.263+ at 24-25 (in

80
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

describing the syntax of the PLUSPTYPE field, explaining that a “subfield” termed

“OPPTYPE” can be present that identifies whether the “Advanced Intra Coding

mode (AIC) mode” is being used), 73 (“The use of [the Advanced Intra Coding

mode] is indicated in the PLUSPTYPE field of the picture header.”); EX1002 at

¶145.)

When the Advanced Intra Coding mode is not applied, H.263+ compresses

the block data (data segments) of all macroblocks (subregions) using the same VLC

table. (EX1002 at ¶146; see Cote at 851 (describing the use of entropy coding in

H.263), 854 (when the Advanced Intra Coding mode is not applied, explaining that

“the standard employs the same VLC table for coding all quantized coefficients”);

H.263+ at 11 (“the [predicted] coefficients are quantized and entropy coded”), 41-

44 (including in Table 16 the VLC table for coefficients)).

But when the Advanced Intra Coding mode is applied, H.263+ compresses

the block data of macroblocks based on two, selectable VLC tables—a first VLC

table that is applied to inter macroblocks, and a second VLC table that is applied to

intra macroblocks. (EX1002 at ¶147; H.263+ at 4 (“Separate Variable Length Code

(VLC) tables are defined for the INTRA blocks”), 74-75 (using a “separate VLC for

INTRA coefficients”).)

Moreover, a POSA would have understood that using different VLC tables

for different macroblock-types (intra and inter) amounts to using different

81
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

compression algorithms for different macroblock-types—one compression

algorithm for each VLC table. (EX1002 at ¶148; see also Cote at 851 (“Entropy

coding [(a type of compression encoding)] is performed by means of variable-

length-codes (VLC’s).”), 854 (“Thus, the advanced intra coding mode employs a

new VLC table for encoding [(compressing)] the quantized coefficients, a table that

is optimized to global statistics of intra macroblocks.”).)

H.263+’s reliance on a different compression algorithm (VLC table) for intra

macroblocks in the Advanced Intra Coding mode optimizes the compression of intra

macroblocks. (EX1002 at ¶149; Cote at 854 (“The main part of the [H.263+]

standard employs the same VLC table for coding all quantized coefficients.

However, this table is designed for inter macroblocks and is not very effective for

coding intra macroblocks.”); see also Cote at 854 (explaining that using different

VLC tables for the intra macroblocks in the Advanced Intra Coding mode “is

optimized to global statistics of intra macroblocks” where “larger coefficients with

smaller runs of zeros are more common”).)

A POSA would have understood that when the Advanced Intra Coding mode

is applied, H.263+ relies on selectable compression algorithms (VLC tables)—one

applied to intra macroblocks and the other applied to inter macroblocks—and where

the macroblock type (intra or inter), as is identified in MCBPC element of the

macroblock header, determines which compression algorithm is selected. (EX1002

82
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

at ¶150; see H.263+ at 33-36 (providing that the macroblock header includes a

“MCBPC” element which defines the “macroblock type” including whether the

macroblock is an “INTRA” or “INTER” type macroblock).) Thus, a POSA would

have understood that the MCBPC element corresponds to a “key code” (as is

claimed) which identifies the compression algorithm applied to compress that

macroblock (“subregion”). (EX1002 at ¶150.)

Further, a POSA would have understood that H.263+ describes a plurality of

compression key code segments (the MCBPC syntax elements—where each

macroblock includes one MCBPC syntax element), each key code segment being

associated with at least one compressed data segment (the macroblocks compressed

data blocks). (EX1002 at ¶151.)

Thus, a POSA would have understood H.263+ as describing “a plurality of

compression key code segments, each key code segment being associated with at

least one compressed data segment.” (EX1002 at ¶152; EX1001 at Claim 10.)

iv. [a plurality of compression key code segments, each

key code segment being associated with at least one
compressed data segment (see limitation iii)] and
identifying a compression algorithm employed to
generate the respective compressed data code from
data for the subregion of the image data stream;
H.263+ discloses or renders obvious this limitation.

83
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

As explained in Section VI.B.2.iii above, a POSA would have understood

H.263+ as describing a plurality of compression key code segments (the MCBPC

syntax element in the macroblock header). (EX1002 at ¶153.)

Additionally, as illustrated below and as explained in Section VI.B.2.iii above

and rehashed below, H.263+ describes the MCBPC syntax element in the

macroblock header as identifying a compression algorithm (VLC table) used to

generate the compressed data block for each macroblock in the picture. (EX1002 at

¶154; H.263+ at Figures 7 (Page 22), 9 (Page 32), and I.1 (Page 74).)

84
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(H.263+ at Figures 7 (Page 22), 8 (Page 22) (annotated to include

MPPTYPE and OPPTYPE subfields), 9 (Page 32), and I.1 (Page 74).)

As described in Section VI.B.2.iii above, H.263+ describes the use of inter

and intra macroblocks. (H.263+ at 33-36; Cote at 852 (“The two switches in Fig. 1

represent the intra/inter mode selection, which is not specified in the standard. Such

a selection is made at the macroblock level.”); EX1002 at ¶155.)

85
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Further, as illustrated above, the MCBPC element of the macroblock header

indicates macroblock type (inter or intra). (H.263+ at 33-36 (providing that the

macroblock header includes a MCBPC element which defines the “macroblock

type” including whether the macroblock is an “INTRA” or “INTER” type

macroblock); EX1002 at ¶155.)

Also as described in Section VI.B.2.iii above, when the Advanced Intra

Coding mode is applied, H.263+ compresses the block data of macroblocks based

on two, selectable VLC tables—a first VLC table that is applied to inter

macroblocks, and a second VLC table that is applied to intra macroblocks. (EX1002

at ¶156; H.263+ at 4 (“Separate Variable Length Code (VLC) tables are defined for

the INTRA blocks”), 74-75 (using a “separate VLC for INTRA coefficients”).)

H.263+’s reliance on a different compression algorithm (VLC table) for intra

macroblocks in the Advanced Intra Coding Mode optimizes the compression of intra

macroblocks. (EX1002 at ¶157; Cote at 854 (“The main part of the [H.263+]

standard employs the same VLC table for coding all quantized coefficients.

However, this table is designed for inter macroblocks and is not very effective for

coding intra macroblocks.”), 854 (explaining that using different VLC tables for the

intra macroblocks in the Advanced Intra Coding Mode “is optimized to global

statistics of intra macroblocks” where “larger coefficients with smaller runs of zeros

are more common”).)

86
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

A POSA would have understood that when the Advanced Intra Coding mode

is applied, H.263+ relies on selectable compression algorithms (VLC tables)—one

applied to intra macroblocks and the other applied to inter macroblocks—and where

the macroblock type (intra or inter), as is identified in MCBPC syntax element of

the macroblock header, determines which compression algorithm is selected.

(EX1002 at ¶158; see H.263+ at 33-36 (providing that the macroblock header

includes a “MCBPC” syntax element which defines the “macroblock type” including

whether the macroblock is an “INTRA” or “INTER” type macroblock).) Thus, a

POSA would have understood that the MCBPC syntax element corresponds to a

“key code” (as is claimed) which identifies the compression algorithm applied to

compress the block data of that macroblock (“subregion”). (EX1002 at ¶158.)

Thus, a POSA would have understood that when the Advanced Intra Coding

mode is being used, H.263+ describes a plurality of compression key code segments

(the MCBPC syntax elements) that identify a compression algorithm being

employed (where one compression algorithm is employed by intra macroblocks, and

another compression algorithm is employed by inter macroblocks) to compress the

block data of the macroblock (subregion). (EX1002 at ¶159.)

Thus, a POSA would have understood H.263+ as describing “key code

segments [that are] identifying a compression algorithm employed to generate the

87
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

respective compressed data code from data for the subregion of the image data

stream.” (EX1002, ¶¶153-160.)

v. a compression header including a cross reference of the

key code segments to the compression algorithms.
H.263+ discloses or renders obvious this limitation.

As explained in Sections VI.B.2.iii-iv above, when the Advanced Intra

Coding mode is used, H.263+ identifies which compression algorithm (VLC table)

is being applied based on the MCBPC syntax element the macroblock header—

where a first compression algorithm (VLC table) is employed to compress intra

macroblocks and a second compression algorithm (VLC table) is employed to

compress inter macroblocks. (EX1002 at ¶162; see H.263+ at 33-36 (providing that

the macroblock header includes a “MCBPC” syntax element which defines the

“macroblock type” including whether the macroblock is an “INTRA” or “INTER”

macroblock); H.263+ at 4 (for the Advanced Intra Coding mode, explaining that

“Separate Variable Length Code (VLC) tables are defined for the INTRA blocks”),

74-75 (for the Advanced Intra Coding mode, using a “separate VLC for INTRA

coefficients”).)

Additionally, H.263+ explains that whether the Advanced Intra Coding mode

is in use is identified in a picture header. Specifically, as illustrated below, when the

Advanced Intra Coding mode is applied, the picture header includes the

88
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

PLUSPTYPE syntax element. (H.263+ at 73 (“The use of [the Advanced Intra

Coding Mode] is indicated in the PLUSPTYPE field of the picture header.”).) The

OPPTYPE subfield in the PLUSPTYPE element identifies whether the Advanced

Intra Coding mode is being applied. (H.263+ at 24 (stating that “Bit 8” of the

OPPTYPE subfield of PLUSPTYPE element identifies whether the “Optional

Advanced Intra Coding (AIC) mode” is being used), 73 (“The use of [the Advanced

Intra Coding mode] is indicated in the PLUSPTYPE field of the picture header.”);

EX1002 at ¶163.)

(H.263+ at Figures 7, 8 (Page 22) (annotated to include MPPTYPE and OPPTYPE

subfields).)

89
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Thus, a POSA would have understood that H.263+ relies on selectable

compression algorithms (VLC tables) when the Advanced Intra Coding mode is

applied, where the OPPTYPE subfield in the PLUSPTYPE syntax element of the

picture header indicates whether the Advanced Intra Coding mode is in use.

(EX1002 at ¶164) Additionally, as explained in Sections VI.B.2.iii-iv above, when

the Advanced Intra Coding mode is in use, the compression algorithm (VLC table)

for each macroblock is selected based on macroblock type, where macroblock type

is identified by the MCBPC syntax element in the macroblock header. (EX1002 at

¶164.) In that way, H.263+ selects compression algorithms based on the MCBPC

syntax element only when the OPPTYPE subfield indicates that the Advanced Intra

Coding mode is in use. (Id.) Thus, the OPPTYPE subfield acts as a “cross reference

of the key code segments [MCBPC syntax elements] to the compression algorithms

[VLC tables]” (as is claimed) because the OPPTYPE subfield indicates (or, in other

words, cross references) whether the MCBPC syntax elements are indicative of the

VLC table to be used for compressing each macroblock. (Id.)

To note, the term “cross reference” as is claimed is not limited by the

embodiments in the specification (see, e.g., ʼ583 Patent at 13:40-45 (“While the

invention may be susceptible to various modifications and alternative forms, specific

embodiments have been shown by way of example in the drawings and have been

described in detail herein. However, it should be understood that the invention is

90
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

not intended to be limited to the particular forms disclosed.”).) Further, what a

“cross reference” comprises was not limited through disclaimer or disavowal during

prosecution of the ʼ583 patent. (See generally EX1003.) A cross reference merely

references information in another place—where, as with H.263+, the OPPTYPE

subfield references information relevant to compression algorithm selection (the

macroblock type) as existing in the MCBPC syntax elements.

Thus, a POSA would have understood the picture header as corresponding to

the “compression header” claimed. (EX1002 at ¶165.) And a POSA would have

understood the picture header as including a cross reference (OPPTYPE subfield in

the PLUSPTYPE syntax element) of the key code segments (MCBPC syntax

elements in the macroblock headers) to the compression algorithms (VLC tables)

being used. (EX1002 at ¶165.) Indeed, when the OPPTYPE subfield indicates that

the Advanced Intra Coding mode is not in use, H.263+ relies on the same

compression algorithm for all macroblocks and, thus, the MCBPC syntax element is

not indicative of compression algorithm type. (EX1002 at ¶165; see Cote at 854

(when the Advanced Intra Coding mode is not in use, explaining that “the [H.263+]

standard employs the same VLC table for coding all quantized coefficients”).) And

when the OPPTYPE subfield indicates that the Advanced Intra Coding mode is in

use, this indication cross references the MCBPC syntax elements as being indicative

91
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

of the compression algorithm that applies (based on macroblock type). (EX1002 at

¶¶161-165.)

Further, while during prosecution the Applicant argued that the Huang

reference only includes a “local header … associated with the each of the

compressed subregions” and “is devoid of a header for the image file” including a

“cross reference of the key code segments to the compression algorithms, as recited

in [original] claim 12,” in contrast, H.263+ clearly includes a “header for the image

file” (the picture header) that cross references (through the OPPTYPE subfield in

the PLUSPTYPE syntax element) the key code segments (the MCBPC syntax

elements in the macroblock headers) to the compression algorithms. (EX1002 at

¶166; EX1003 at 275; see also EX1003 at 284 (the Examiner similarly finding that

Claim 10 is allowable based on not finding prior art that teaches “a compression

header including a cross reference of the key code segments to the compression

algorithms” where, H.263+—which was not considered during prosecution—

clearly requires such).)

Thus, a POSA would have understood H.263+ as describing “a compression

header including a cross reference of the key code segments to the compression

algorithms.” (EX1002, ¶¶161-167.)

92
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

VII. SUBSTANTIAL NEW QUESTION OF PATENTABILITY #2 – NON-

STATUTORY OBVIOUSNESS-TYPE DOUBLE PATENTING

A. Nonstatutory Obviousness-Type Double Patenting

1. Legal Standard
35 U.S.C. § 101 requires that “[w]hoever invents or discovers any new and

useful process, machine, manufacture, composition of matter, or any new and useful

improvement thereof, may obtain a patent therefor, subject to the conditions and

requirements of this title.” “Thus, § 101 forbids an individual from obtaining more

than one patent on the same invention, i.e., double patenting.” AbbVie Inc. v.

Mathilda & Terrence Kennedy Institute of Rheumatology Trust, 764 F.3d 1366, 1372

(Fed. Cir. 2014).

35 U.S.C. § 101 is the basis for statutory double patenting rejections. In turn,

the non-statutory double patenting rejection is based on a judicially created doctrine

grounded in public policy intended “to prevent the possibility of multiple suits

against an accused infringer by different assignees of patents claiming patentably

indistinct variations of the same invention.” (MPEP § 804 (citing In re Van

Ornum, 686 F.2d 937, 944-48 (CCPA 1982).) “A nonstatutory double patenting

rejection is appropriate where the conflicting claims are not identical, but at least

one examined application claim is not patentably distinct from the reference claim(s)

because the examined application claim is either anticipated by, or would have been

93
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

obvious over, the reference claim(s).” (MPEP § 804.II.B (citing In re Berg, 140 F.3d

1428 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 1052 (Fed. Cir. 1993)).)

“Before consideration can be given to the issue of double patenting [including

nonstatutory double patenting], two or more patents or applications must have at

least one common inventor, common applicant, and/or be commonly

assigned/owned[.]” (MPEP § 804.) Thus, a patent qualifies as a reference for a

nonstatutory double patenting rejection if that patent “ha[s] at least one common

inventor, common applicant, and/or [is] commonly assigned/owned” as the patent

being challenged. (MPEP § 804.)

“A nonstatutory double patenting rejection, if not based on an anticipation

rationale …, is ‘analogous to [a failure to meet] the nonobviousness requirement

of 35 U.S.C. § 103’ except that the patent disclosure principally underlying the

double patenting rejection [—the double patenting reference patent—] is not

considered prior art.” (MPEP § 804.II.B.2 (citing In re Braithwaite, 379 F.2d 594

(CCPA 1967)).) “The analysis employed with regard to nonstatutory double

patenting is similar to, but not necessarily the same as that undertaken under 35 USC

§ 103.” (MPEP § 804.II.B.2 (citing In re Braat, 937 F.2d 589, 592-93 (Fed. Cir.

1991) (citing In re Longi, 759 F.2d 887, 892 n.4 (Fed. Cir. 1985))).) “In view of the

similarities, the factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1,

148 USPQ 459 (1966) that are applied for establishing a background for determining

94
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

obviousness under 35 U.S.C. § 103 should typically be considered when making a

nonstatutory double patenting analysis based on ‘obviousness.’” (MPEP §

804.II.B.2.) A nonstatutory double patenting rejection based on obviousness is often

termed an obviousness-type double patenting rejection (OTDP rejection).

Further, when determining whether a challenged patent’s claims would have

been obvious over a double patenting reference patent’s claims, an OTDP rejection

can rely on the double patenting reference patent’s claims in view of prior art

references. See, e.g., Otsuka Pharm. Co. v. Sandoz, Inc., 678 F.3d 1280, 1296–97

(Fed. Cir. 2012); In re Ockert, 245 F.2d 467, 469 (CCPA 1957) (“It is also well

settled that, in determining whether the claims of an application are patentably

distinct from those of a patent it is proper to consider what is disclosed by the prior

art.”)

Moreover, “[i]f the application under examination is the later-filed

application, or both applications are filed on the same day, only a one-way

determination of distinctness is needed in resolving the issue of double patenting,

i.e., whether the invention claimed in the application would have been anticipated

by, or an obvious variation of, the invention claimed in the patent.” (MPEP

§ 804.B.II.b (citing In re Berg, 140 F.3d 1438, 46 USPQ2d 1226 (Fed. Cir. 1998)

95
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(the court applied a one-way test where both applications were filed the same day)).) 4

For example, in determining the distinctness of claims in an OTDP analysis, it has

been held that there may not be “a patentable distinction between [a] method of using

[a] device and [a] device itself.” (In re Lonardo, 119 F.3d 960, 968 (Fed. Cir. 1997)

(“Restorative Care argues that the method of using the device would not have been

obvious over a claim to the device. We do not agree that there is a patentable

distinction between the method of using the device and the device itself. The claimed

structure of the device suggests how it is to be used and that use thus would have

been obvious.”).)

Patent Owners can overcome non-statutory double patenting rejections by

filing terminal disclaimers before the reference patent expires. Before the reference

patent expires, “the submission of a terminal disclaimer in compliance with 37 CFR

1.321(c) or (d) to overcome a double patenting rejection ensures that a patent owner

with multiple patents claiming obvious variations of one invention retains all those

patents or sells them as a group”—thereby preventing the possibility of harassing

4
The one-way test applies to this Request because the challenged ʼ583 Patent and

the double-patenting reference patents (the ʼ151 and ʼ639 Patents) were filed on the

same day. A two-way test also exists, but only should be used in other circumstances

and is not relevant to this Request.

96
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

accused infringers with multiple suits “by different assignees of patents claiming

patentably indistinct variations of the same invention.” (MPEP § 804 (citing In re

Van Ornum, 686 F.2d at 944-45).) However, a terminal disclaimer filed after the

expiration of the reference patent is not effective to overcome a non-statutory double

patenting rejection. (See Boehringer Ingelheim Intʼl v. Barr Laboratories, 592 F.3d

1340 (Fed. Cir. 2010); MPEP § 804.02.)

2. Obviousness-Type Double Patenting in Reexamination

Proceedings
As indicated in the MPEP, nonstatutory double patenting rejections can raise

substantial new questions of patentability in reexamination proceedings. (See MPEP

§ 2258.I.D.) And while “reexamination proceedings are based on ‘prior art’ patents

and printed publications[,] … there are certain exceptions, even for reexaminations

ordered under 35 U.S.C. § 304.” (MPEP § 2258.I.) Particularly, in In re Lonardo,

“the Federal Circuit upheld a nonstatutory double patenting rejection in which the

patent upon which the rejection was based and the patent under reexamination

shared the same effective filing date.” (MPEP § 2258.I (citing In re Lonardo, 119

F.3d 960 (Fed. Cir. 1997); see also MPEP § 2258.I.D (“Accordingly, the issue of

double patenting, over prior art patents or non-prior art patents, is appropriate for

consideration in reexamination under 35 U.S.C. § 302, both as a basis for ordering

reexamination under 35 U.S.C. § 304 and during subsequent examination on the

97
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

merits. The issue of double patenting is to be considered by the examiner when

making the decision on the request for reexamination. The examiner should

determine whether the issue of double patenting raises a substantial new question of

patentability. The issue of double patenting is also to be considered during the

examination stage of reexamination proceeding. In the examination stage, the

examiner should determine whether a rejection based on double patenting is

appropriate.”); In re Lonardo, 119 F.3d 960, 966 (Fed. Cir. 1997) (“Since the statute

in other places refers to prior art in relation to reexamination[,] it seems apparent

that Congress intended that the phrases ‘patents and publications’ and ‘other patents

or printed publications’ in section 303(a) not be limited to prior art patents or printed

publications.”).)

As shown below, and as further explained by Dr. Joseph Havlicek,

nonstatutory obviousness-type double patenting raises substantial new questions of

patentability with regard to Claims 1 and 10 of the ʼ583 patent. (EX1002, ¶¶174-

204.)

B. Overview of Patents Qualifying as Nonstatutory Double Patenting

References

1. U.S. Patent 6,792,151 (Ex. 1010)

U.S. Patent 6,792,151 (the “ʼ151 Patent” (EX1010)) was filed on November

24, 1999 and issued on September 14, 2004. (EX1010 at Cover.) The ʼ151 Patent

98
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

and ʼ583 Patent share the same filing date—November 24, 1999. (EX1001 at

Cover; EX1010 at Cover.) The ʼ151 Patent and ʼ583 Patent both name Robert D.

Barnes and Robert C. Gemperline as inventors, and the General Electric Company

as the assignee. (Id.) Thus, as explained in Section VII.A.1 above, the ʼ151 Patent

qualifies as a double patenting reference for the ʼ583 Patent. (See MPEP §§ 804,

2258.I.)

As to content, the ʼ151 Patent and ʼ583 Patent share exactly the same figures

(compare EX1001 at Figures 1–13 with EX1010 at Figures 1–13) and nearly

identical specifications.

Moreover, the reference claims of the ʼ151 Patent relied on in the proposed

OTDP rejection below include claim 1 (independent), and claims 6, 7, 8, 9, 10

(dependent on claim 1). These claims describe a method for compressing image data

and require:

1. A method for compressing image data from an uncompressed

image data stream, the method comprising the steps of:

(a) receiving the uncompressed image data stream;

(b) analyzing the uncompressed image data stream by

contiguous subregions to identify from a plurality of compression
algorithms a compression algorithm that permits optimal lossless
compression of each subregion;

(c) selecting an optimal data compression algorithm for each

subregion based upon the analysis performed in step (b); and

99
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

(d) compressing the image data stream by subregion based upon

the selected algorithms.

6. The method of claim 1, wherein different compression

algorithms are selected for subregions having different entropy levels.

7. The method of claim 6, wherein the entropy levels are

determined by analysis of variations between values of individual
picture elements defined by the image data stream for each subregion.

8. The method of claim 7, wherein the variations are computed

based upon a predictor algorithm analyzing data representative of
each individual picture element and data representative of picture
elements neighboring the respective picture element.

9. The method of claim 8, wherein the predictor algorithm is

selected from a set of candidate predictor algorithms based upon
portions of the image data stream descriptive of the image.

10. The method of claim 9, wherein the portions of the image

data stream descriptive of the image are included in an image header
portion appended to an image defining portion of the image data
stream.

2. U.S. Patent 7,050,639 (Ex. 1011)

U.S. Patent 7,050,639 (the “ʼ639 Patent” (EX1011)) was filed on November

24, 1999 and issued on May 23, 2006. (EX1011 at Cover.) The ʼ639 Patent and

ʼ583 Patent share the same filing date—November 24, 1999. (EX1001 at

Cover; EX1011 at Cover.) The ʼ639 Patent and ʼ583 Patent both name Robert D.

Barnes and Robert C. Gemperline as inventors, and the General Electric Company

as the assignee. (Id.) Thus, as explained in Section VII.A.1 above, the ʼ639 Patent

qualifies as a double patenting reference for the ʼ583 Patent. (See MPEP §§ 804,

2258.I.)

100
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

As to content, the ʼ639 Patent and ʼ583 Patent share exactly the same figures

(compare EX1001 at Figures 1–13 with EX1011 at Figures 1–13) and nearly

identical specifications.

Moreover, the reference claims of the ʼ639 Patent relied on in the proposed

OTDP rejection below include claim 18 (independent), and claim 23 (dependent on

claim 18). These claims describe a system for storing, transmitting, and viewing

compressed images and require:

18. A system for storing, transmitting and viewing images, the system
comprising:

a data compression station configured to store a plurality of

compression code tables for conversion of uncompressed image data
representative of individual picture elements to lossless compressed
image data and to execute a compression routine in which an image
data stream is converted to a compressed file by dividing into
subregions and each subregion compressing in accordance with a
compression code table selected from the plurality of compression
code tables based upon which compression code table provides
optimal lossless compression of the subregion;

a data storage device for receiving and storing the compressed

file; and

an image decompression station configured to store the

plurality of compression code tables, to access the compressed file
from the data storage device, and to execute a decompression routine
in which the compression code tables applied to compress the image
data stream are applied to decompress the compressed file to
reconstruct the image data stream.

23. The system of claim 18, wherein the compression routine

includes encoding of identifiers of the selected compression code
tables within the compressed file, and wherein the decompression

101
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

routine includes analysis of the identifiers for selection of the same

compression code tables for decompression of the compressed file.

C. Obviousness-Type Double Patenting raises a substantial new

question of patentability for Claims 1 and 10.
As explained above in Sections VII.A-B, the ʼ151 and ʼ639 Patents qualify

as Obviousness-Type Double Patenting (OTDP) references for the ʼ583 Patent—the

ʼ151 and ʼ639 Patents share the same filing date, inventors, and assignee. (EX1001

at Cover; EX1010 at Cover; EX1011 at Cover; see MPEP §§ 804, 2258.I.)

During prosecution of the ʼ583 Patent, the Examiner completed an Inventor

Name Search on the Palm Intranet, where those search results included the patent

applications that became the ʼ151 and ʼ639 Patents. (EX1003 at 11.) However, the

Examiner never raised any issues related to double patenting during prosecution.

(See generally EX1003.)

And as shown below, the ʼ151 and ʼ639 Patents in view of H.263+ render

Claims 1 and 10 of the ʼ583 Patent invalid under OTDP.

Particularly, while the Examiner allowed the ʼ583 Patent’s Claim 1 following

an amendment that required a selected predictor algorithm being identified in a

compression header, the ʼ151 Patent’s claims 1 (independent), 6, 7, 9, and 10 (where

claims 6, 7, 9, and 10 are dependent on claim 1) in view of H.263+ plainly requires

such.

102
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Additionally, while the Examiner allowed the ʼ583 Patent’s Claim 10

following an amendment that required that the compressed image file include a

compression header including a cross reference of the key code segments to the

compression algorithm, the ʼ639 Patent’s claims 18 (independent) and 23 (dependent

on claim 18) in view of H.263+ plainly requires such.

1. Proposed Rejection 1: The ʼ151 Patent’s Claim 10 (which also

includes the limitations of Claims 1 and 6-9) in view of H.263+
renders the ʼ583 Patent’s Claim 1 invalid based on obviousness-
type double patenting
Claim 1 of the ʼ583 Patent is a patentably indistinct, obvious variant of claim

10 of the ʼ151 Patent in view of H.263+. Claim 10 of the ’151 Patent includes the

limitations of Claims 1 and 6-9.5 (See MPEP § 804.B.II.b.) As shown below and

supported by the declaration of Dr. Joseph Havlicek (EX1002 at ¶¶174-190), the

challenged Claim 1 of the ʼ583 Patent and the reference Claim 10 (including the

limitations of claims 1 and 6-9) of the ʼ151 Patent both describe methods of

compressing image data, and any lack of direct overlap between these claims is

rendered obvious by the teachings of H.263+.

5
In this Request, “Claim 10” of the ʼ151 Patent refers to Claim 10 on its own and

the limitations of claims 1 and 6-9 which Claim 10 includes.

103
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

i. Motivation to Combine Claim 10 (which also includes

the limitations of claims 1 and 6-9) of the ʼ151 Patent
and H.263+
A POSA would have been motivated to combine the disclosure of Claims 10

(including the limitations of claims 1 and 6-9) of the ʼ151 Patent with the teachings

of H.263+. Specifically, Claim 10 of the ʼ151 Patent and H.263+ both describe “a

method of compressing image data.” (EX1010 at Claim 1; H.263+ at Cover

Summary (PDF Page 3) (explaining that H.263+ standard provides “improved

compression performance” for the encoded images that comprise the video data), 4

(explaining that the Advanced Intra Coding mode of Annex I “significantly

improves the compression performance over the INTRA coding of the core H.263

syntax”); 7 (“The [H.263] source code operates on non-interlaced pictures….”).)

Thus, a POSA would have been motivated to modify the method of compressing

image data described in Claim 10 of the ʼ151 Patent with additional insight from and

informed by H.263+’s further teachings regarding compressing image data.

(EX1002 at ¶174.)

In addition, a POSA would have been motivated to combine Claim 10 of the

ʼ151 Patent and H.263+ because both references present an image compression

method, where H.263+ provides greater detail into how an image compression

method can be applied. (EX1002 at ¶175.)

104
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

For example, Claim 10 of the ʼ151 Patent does not explicitly require

identifying the selected compression algorithm using a key code that is compiled

with the data for its respective subregion (as is required by ʼ583 Claim 1). Thus, a

POSA would have been motivated to consider H.263+’s teaching of identifying a

selected compression algorithm using a key code that is compiled with the data for

its respective subregion (see supra Section VI.B.1.v; EX1002 at ¶¶161-167) and

apply this teaching to Claim 10 of the ʼ151 Patent. (EX1002 at ¶176.)

This discussion demonstrates that several rationales support a motivation to

combine in this obviousness analysis:

x Combing prior art elements (the image compression methods of H.263+)

according to known methods (compressing an image file, as taught by Claim

10 of the ʼ151 Patent and H.263+) to yield predictable results (building on the

disclosure of the image compression method of Claim 10 of the ʼ151 Patent

with the further image compression features of H.263+) (EX1002 at ¶177);

x Use of a known technique (H.263+’s image compression method) to improve

similar devices (the image compression method of Claim 10 of the ʼ151

Patent) in the same way (H.263 and Claim 10 of the ʼ151 Patent both describe

image compression methods) (EX1002 at ¶178); and

x Applying a known technique (H.263+’s image compression method) to a

known device (the image compression method of Claim 10 of the ʼ151 Patent)

105
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

ready for improvement to yield predictable results (using the image

compression method of H.263+ to predictably build on/improve the image

compression method of Claim 10 of the ʼ151 Patent) (EX1002 at ¶179).

KSR Intʼl Co. v. Teleflex Inc., 550 U.S. 398, 415-22 (2007). Indeed, “[t]he

combination of familiar elements according to known methods is likely to be

obvious when it does no more than yield predictable results.” (KSR Intʼl Co., 550

U.S. at 416; see also M.P.E.P. § 2143(I)(A); see also EX1002, ¶¶177-179.)

Additionally, because the combination of Claim 10 of the ʼ151 Patent and

H.263+ involves the simple combination of elements that would yield predictable

results, a POSA would also have had a reasonable expectation of success in making

this obvious combination. (EX1002 at ¶180.) Particularly, Claim 10 of the ʼ151

Patent and H.263+ both teach image compression methods that use similar elements,

and both select prediction and compression algorithms that are used to code image

data. (EX1010 at Claims 1 and 6-10; H.263+ at 73-84 (Annex I, Advanced Intra

Coding mode).) Claim 10 of the ʼ151 Patent explicitly discloses an image

compression method and a POSA would have understood that H.263+ teachings

further related to an image compression method could be implemented into the

broadly presented image compression method of Claim 10 of the ʼ151 Patent.

(EX1002, ¶180.) This would require minimal modifications to the image

compression method of Claim 10 of the ʼ151 Patent—particularly because this

106
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

image compression method is so broad when considering the scope of the claims—

and the combination would have led to a predictable result (further applying the

image compression features as described in H.263+). (EX1002, ¶180; see Pfizer,

Inc. v. Apotex, Inc., 480 F.3d 1348, 1364 (Fed. Cir. 2007).)

Accordingly, a POSA would have been motivated to combine the teachings

of Claim 10 of the ʼ151 Patent in view of H.263+. (EX1002 at ¶¶174-181.)

ii. Claim 10 (which also includes the limitations of claims

1 and 6-9) of the ʼ151 Patent in view of H.263+ renders
Claim 1 of the ʼ583 invalid under OTDP
The proposed OTDP rejection is presented in the table below—where the ʼ583

Patent’s Claim 1’s limitations are presented in the left column, and the obvious

variants thereof from Claim 10 of the ʼ151 Patent are provided in the right column.

Further, in view of the motivation to combine presented in Section VII.C.1.i above,

this table presents that any limitations that may be considered as non-obvious

variants of Claim 10 of the ʼ151 Patent are obvious further in view of H.263+.

(EX1002 at ¶182.)

’583 Patent Claim Exemplary citations to the ʼ151 Patent’s Claim 10

Language (which also includes the limitations of claims 1 and 6-9)

[1Pre] A method “[ʼ151 Claim 1] A method for compressing image data

for compressing an from an uncompressed image data stream, the method
image data stream, comprising the steps of: …” (EX1010 at Claim 1.)
the method
comprising the A POSA would have understood that the ʼ583 Patent’s
steps of: preamble ([1PRE]) is an indistinct, obvious variant of the

107
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

’583 Patent Claim Exemplary citations to the ʼ151 Patent’s Claim 10

Language (which also includes the limitations of claims 1 and 6-9)

language of the ʼ151 Patent’s Claim 1 quoted above.

(EX1002 at ¶183.)

[1A] (a) dividing “[ʼ151 Claim 1] … (b) analyzing the uncompressed image
the image data data stream by contiguous subregions to identify from a
stream into a plurality of compression algorithms a compression
plurality of algorithm that permits optimal lossless compression of each
subregions, subregion; …” (EX1010 at Claim 1.)

A POSA would have understood that the ʼ583 Patent’s

limitation [1A] is an indistinct, obvious variant of the
language of the ʼ151 Patent’s Claim 1 quoted above.
(EX1002 at ¶184.)

[1B-1] (b) selecting “[ʼ151 Patent Claims 8, 9, and 10]

a predictor
algorithm from a 8. The method of claim 7, wherein the variations are
plurality of computed based upon a predictor algorithm analyzing
predictor data representative of each individual picture element and
algorithms to data representative of picture elements neighboring the
generate predicted respective picture element.
pixel data values
9. The method of claim 8, wherein the predictor
…
algorithm is selected from a set of candidate predictor
algorithms based upon portions of the image data stream
descriptive of the image. (EX1010 at Claims 8-9.)

A POSA would have understood that the ʼ583 Patent’s

limitation [1B-1] is an indistinct, obvious variant of the
language of the ʼ151 Patents Claims 8-9 quoted above.
(EX1002 at ¶185.)

[1B-2] and “[ʼ151 Patent Claim 1] (b) analyzing the uncompressed

compressing data image data stream by contiguous subregions to identify
of each subregion from a plurality of compression algorithms a compression
in accordance with algorithm that permits optimal lossless compression of each
one of a plurality subregion;

108
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

’583 Patent Claim Exemplary citations to the ʼ151 Patent’s Claim 10

Language (which also includes the limitations of claims 1 and 6-9)

of compression (c) selecting an optimal data compression algorithm for

algorithms; each subregion based upon the analysis performed in step
(b); …” (EX1010 at Claim 1.)

A POSA would have understood that the ʼ583 Patent’s

limitation [1B-2] is an indistinct, obvious variant of the
language of the ʼ151 Patent’s Claim 1 quoted above.
(EX1002 at ¶186.)

[1C-1] (c) [ʼ151 Patent Claim 1] A method for compressing image

compiling a data from an uncompressed image data stream, the method
compressed data comprising the steps of:
file including
compressed data (a) receiving the uncompressed image data stream;
for each subregion
(b) analyzing the uncompressed image data stream by
and key code
contiguous subregions to identify from a plurality of
identifying the
compression algorithms a compression algorithm that
compression
permits optimal lossless compression of each subregion;
algorithms applied
to compress each (c) selecting an optimal data compression algorithm for
subregion, … each subregion based upon the analysis performed in step
(b); and

(d) compressing the image data stream by subregion based

upon the selected algorithms.

The ʼ151 Patent’s Claims do not explicitly require

identifying the selected compression algorithm using a key
code that is compiled with the data for its respective
subregion.

However, as explained in Section VI.B.1.v above, a POSA

would have understood H.263+ as teaching identifying
selected compression algorithms using a key code that is
compiled with the data for its respective subregion.
(EX1002 at ¶187; see also EX1002 at ¶¶121-125.)

109
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

’583 Patent Claim Exemplary citations to the ʼ151 Patent’s Claim 10

Language (which also includes the limitations of claims 1 and 6-9)

Additionally, a POSA would have been motivated to

combine Claim 1 of the ʼ151 Patent with H.263+ in this
way based on the motivation to combine presented in
Section VII.C.1.i above. (EX1002 at ¶187; see EX1002 at
¶¶174-181.)

Thus, when considering Claim 1 of the ʼ151 Patent in view

of H.263+, a POSA would have understood that the ʼ583
Patent’s limitation [1C-1] is an indistinct, obvious variant.
(EX1002 at ¶187.)

[1C-2] wherein the The ʼ151 Patent’s Claims 1 and 6-10 do not explicitly
compressed data teach wherein the compressed data comprises pixel data
comprises pixel that includes predictor error data representative of
data that includes differences between predicted pixel data values and actual
predictor error data pixel data values
representative of
differences However, as explained in Section VI.B.1.vi above, a POSA
between predicted would have understood H.263+ as teaching wherein the
pixel data values compressed data comprises pixel data that includes
and actual pixel predictor error data representative of differences between
data values, and predicted pixel data values and actual pixel data values.
(EX1002 at ¶188; see also EX1002 at ¶¶126-128.)

Additionally, a POSA would have been motivated to

combine Claim 10 of the ʼ151 Patent with H.263+ in this
way based on the motivation to combine presented in
Section VII.C.1.vi above. (EX1002 at ¶188; see EX1002
at ¶¶174-181.)

Thus, when considering Claim 10 of the ʼ151 Patent in

view of H.263+, a POSA would have understood that the
ʼ583 Patent’s limitation [1C-2] is an indistinct, obvious
variant. (EX1002 at ¶188.)

[1D] (d) appending [ʼ151 Patent at Claims 8, 9, and 10]

a compression
header to the

110
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

’583 Patent Claim Exemplary citations to the ʼ151 Patent’s Claim 10

Language (which also includes the limitations of claims 1 and 6-9)

compressed data 8. The method of claim 7, wherein the variations are

file, the computed based upon a predictor algorithm analyzing
compression data representative of each individual picture element and
header including data representative of picture elements neighboring the
predictor data respective picture element.
identifying the
selected predictor 9. The method of claim 8, wherein the predictor
algorithm used to algorithm is selected from a set of candidate predictor
generate the algorithms based upon portions of the image data stream
predicted pixel data descriptive of the image.
values.
10. The method of claim 9, wherein the portions of the
image data stream descriptive of the image are included in
an image header portion appended to an image defining
portion of the image data stream.

While Claims 8-9 of the ʼ151 Patent teach the selection of a

prediction algorithm and Claim 10 of the ʼ151 Patent
teaches the appending to the compressed data an image
header, the ʼ151 Patent’s Claim 10 does not explicitly
require the image header to identify the selected predictor
algorithm.

However, as explained in Section VI.B.1.vii above, a

POSA would have understood H.263+ as teaching the
image header as identifying the selected predictor
algorithm. (EX1002 at ¶189; see also EX1002 at ¶¶129-
133.)

Additionally, a POSA would have been motivated to

combine the ʼ151 Patent’s Claim 10 with H.263+ in this
way based on the motivation to combine presented in
Section VII.C.1.i above. (EX1002 at ¶189; see EX1002 at
¶¶174-181.)

Thus, when considering Claim 10 of the ʼ151 Patent in

view of H.263+, a POSA would have understood that the

111
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

’583 Patent Claim Exemplary citations to the ʼ151 Patent’s Claim 10

Language (which also includes the limitations of claims 1 and 6-9)

ʼ583 Patent’s limitation [1D] is an indistinct, obvious

variant. (EX1002 at ¶189.)

Thus, for the reasons provided above, Claim 1 of the ʼ583 Patent is a

patentably indistinct, obvious variant of Claim 10 of the ʼ151 Patent in view of

H.263+. (EX1002 at ¶¶174-190.)

2. Proposed Rejection 2: The ʼ639 Patent’s Claim 23 (which also

includes the limitations of Claim 18) in view of H.263+ renders
the ʼ583 Patent’s Claim 10 invalid based on obviousness-type
double patenting
Claim 10 of the ʼ583 Patent is a patentably indistinct, obvious variant of Claim

23 of the ʼ639 Patent in view of H.263+. (See MPEP § 804.B.II.b) Claim 23 of the

ʼ639 Patent includes the limitations of Claim 18.6 As shown below and supported

by the declaration of Dr. Joseph Havlicek (EX1002 at ¶¶191-204), the challenged

Claim 10 of the ʼ583 Patent and the reference Claim 23 of the ʼ639 Patent both

describe image compression, and any lack of direct overlap between these claims is

rendered obvious by the teachings of H.263+.

6
In this Request, “Claim 23” of the ʼ639 Patent refers to Claim 23 on its own and

the limitations of Claim 18 which Claim 23 includes.

112
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

i. Motivation to Combine Claim 23 (which also includes

the limitations of Claim 18) of the ʼ639 Patent and
H.263+
A POSA would have been motivated to combine the disclosure of Claim 23

of the ʼ639 Patent with the teachings of H.263+. Specifically Claim 23 of the ʼ639

Patent and H.263+ both describe image compression. (EX1011 at Claim 18; H.263+

at Cover Summary (PDF Page 3) (explaining that H.263+ standard provides

“improved compression performance” for the encoded images that comprise the

video data), 4 (explaining that the Advanced Intra Coding mode of Annex I

“significantly improves the compression performance over the INTRA coding of the

core H.263 syntax”); 7 (“The [H.263] source code operates on non-interlaced

pictures….”).) Thus, a POSA would have been motivated to modify the image

compression described in Claim 23 of the ʼ639 Patent with additional insight from

and informed by H.263+’s further teachings regarding compressing image data.

(EX1002 at ¶191.)

In addition, a POSA would have been motivated to combine Claim 23 of the

ʼ639 Patent and H.263+ because both references describe image compression, where

H.263+ provides greater detail into how image compression can be applied.

(EX1002 at ¶192.) For example, Claim 23 of the ʼ639 Patent does not explicitly

require “a compression header including a cross reference of the key code segments

to the compression algorithms.” (ʼ583 Patent at Claim 10.) Thus, a POSA would

113
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

have been motivated to consider H.263+’s teaching that a compression header

includes a cross reference of the key code segments to the compression algorithms

(see supra Section VI.B.2.v; EX1002 at ¶¶161-167) and apply this teaching to Claim

23 of the ʼ639 Patent. (EX1002 at ¶193.)

This discussion demonstrates that several rationales support a motivation to

combine in this obviousness analysis:

x Combing prior art elements (the image compression methods of H.263+)

according to known methods (compressing an image file, as taught by Claim

23 of the ʼ639 Patent and H.263+) to yield predictable results (building on the

disclosure of image compression of Claim 23 of the ʼ639 Patent with the

further image compression features of H.263+) (EX1002 at ¶194);

x Use of a known technique (H.263+’s image compression) to improve similar

devices (the image compression of Claim 23 of the ʼ639 Patent) in the same

way (H.263+ and Claim 23 of the ʼ639 Patent both describe image

compression) (EX1002 at ¶195); and

x Applying a known technique (H.263+’s image compression) to a known

device (the image compression of Claim 23 of the ʼ639 Patent) ready for

improvement to yield predictable results (using the image compression of

H.263+ to predictably build on/improve the image compression described in

Claim 23 of the ʼ639 Patent) (EX1002 at ¶196).

114
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

KSR Intʼl Co. v. Teleflex Inc., 550 U.S. 398, 415-22 (2007). Indeed, “[t]he

combination of familiar elements according to known methods is likely to be

obvious when it does no more than yield predictable results.” (KSR Intʼl Co., 550

U.S. at 416; see also M.P.E.P. 2143(I)(A); see also EX1002, ¶¶194-196.)

Additionally, being that the combination of Claim 23 of the ʼ639 Patent and

H.263+ involves the simple combination of elements that would yield predictable

results, a POSA would also have had a reasonable expectation of success in making

this obvious combination. (EX1002 at ¶197.) Particularly, Claim 23 of the ʼ639

Patent and H.263+ both describe image compression techniques that use similar

elements, and both select compression algorithms that are used to code image data.

(EX1011 at Claim 23; H.263+ at 73-84 (Annex I, Advanced Intra Coding mode).)

Claim 23 of the ʼ639 Patent explicitly discloses image compression and a POSA

would have understood that H.263+’s teachings further related to image

compression could be implemented into the broadly presented image compression

technique of Claim 23 of the ʼ639 Patent. (EX1002 at ¶197.) This would require

minimal modifications to the image compression of Claim 23 of the ʼ639 Patent—

particularly because this image compression is so broad when considering the scope

of the claims—and the combination would have led to a predictable result (further

applying the image compression features as described in H.263+). (EX1002 at ¶197;

see Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348, 1364 (Fed. Cir. 2007).)

115
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Accordingly, a POSA would have been motivated to combine the teachings

of Claim 23 of the ʼ639 Patent and H.263+. (EX1002 at ¶¶191-198.)

ii. Claim 23 (which also includes the limitations of Claim

18) of the ʼ639 Patent in view of H.263+ renders Claim
10 of the ʼ583 invalid under OTDP
The proposed OTDP rejection is presented in the table below—where the ʼ583

Patent’s Claim 10’s limitations are presented in the left column, and the obvious

variants thereof from Claim 23 of the ʼ639 Patent are provided in the right column.

Further, in view of the motivation to combine presented in Section VII.C.2.i above,

this table presents that any limitations that may be considered as non-obvious

variants of Claim 23 of the ʼ639 Patent are obvious further in view of H.263+.

(EX1002 at ¶199.)

’583 Patent ʼ639 Patent’s Claim 23 (which also includes the

Claim Language limitations of Claim 18)

[10Pre] A “[ʼ639 Patent Claim 23]

compressed
image data file 18. A system for storing, transmitting and viewing images,
comprising: the system comprising:

a data compression station configured to store a plurality of

compression code tables for conversion of uncompressed
image data representative of individual picture elements to
lossless compressed image data and to execute a
compression routine in which an image data stream is
converted to a compressed file by dividing into subregions
and each subregion compressing in accordance with a
compression code table selected from the plurality of
compression code tables based upon which compression

116
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

’583 Patent ʼ639 Patent’s Claim 23 (which also includes the

Claim Language limitations of Claim 18)

code table provides optimal lossless compression of the

subregion;

a data storage device for receiving and storing the

compressed file; and

an image decompression station configured to store the

plurality of compression code tables, to access the
compressed file from the data storage device, and to execute
a decompression routine in which the compression code
tables applied to compress the image data stream are applied
to decompress the compressed file to reconstruct the image
data stream.

23. The system of claim 18, wherein the compression routine

includes encoding of identifiers of the selected compression
code tables within the compressed file, and wherein the
decompression routine includes analysis of the identifiers for
selection of the same compression code tables for
decompression of the compressed file.” (EX1011 at Claims
18, 23.)

A POSA would have understood that the preamble of Claim

10 of the ʼ583 Patent ([10PRE]) is an indistinct, obvious
variant of the language of the ʼ639 Patent’s Claim 23 quoted
above. (EX1002 at ¶200.)

[10A] a plurality “[ʼ639 Patent Claim 18]

of compressed
data segments, 18. A system for storing, transmitting and viewing images,
each segment the system comprising:
including
a data compression station configured to store a plurality of
compressed data
compression code tables for conversion of uncompressed
code for a
image data representative of individual picture elements to
subregion of an
lossless compressed image data and to execute a
image data
compression routine in which an image data stream is
stream;
converted to a compressed file by dividing into

117
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

’583 Patent ʼ639 Patent’s Claim 23 (which also includes the

Claim Language limitations of Claim 18)

subregions and each subregion compressing in accordance

with a compression code table selected from the plurality of
compression code tables based upon which compression
code table provides optimal lossless compression of the
subregion;

a data storage device for receiving and storing the

compressed file; and

an image decompression station configured to store the

plurality of compression code tables, to access the
compressed file from the data storage device, and to execute
a decompression routine in which the compression code
tables applied to compress the image data stream are applied
to decompress the compressed file to reconstruct the image
data stream.” (EX1011 at Claim 18.)

A POSA would have understood that the ʼ583 Patent’s

limitation [10A] is an indistinct, obvious variant of the
language of the ʼ639 Patent’s Claim 18 quoted above.
(EX1002 at ¶201.)

[10B] a plurality “[ʼ639 Patent Claim 23]

of compression
key code 18. A system for storing, transmitting and viewing images,
segments, each the system comprising:
key code segment
a data compression station configured to store a plurality of
being associated
compression code tables for conversion of uncompressed
with at least one
image data representative of individual picture elements to
compressed data
lossless compressed image data and to execute a
segment and
compression routine in which an image data stream is
identifying a
converted to a compressed file by dividing into subregions
compression
and each subregion compressing in accordance with a
algorithm
compression code table selected from the plurality of
employed to
compression code tables based upon which compression
generate the
respective

118
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

’583 Patent ʼ639 Patent’s Claim 23 (which also includes the

Claim Language limitations of Claim 18)

compressed data code table provides optimal lossless compression of the

code from data subregion;
for the subregion
of the image data a data storage device for receiving and storing the
stream; compressed file; and

an image decompression station configured to store the

plurality of compression code tables, to access the
compressed file from the data storage device, and to execute
a decompression routine in which the compression code
tables applied to compress the image data stream are applied
to decompress the compressed file to reconstruct the image
data stream.

23. The system of claim 18, wherein the compression routine

includes encoding of identifiers of the selected
compression code tables within the compressed file, and
wherein the decompression routine includes analysis of the
identifiers for selection of the same compression code
tables for decompression of the compressed file.”
(EX1011 at Claims 18, 23.)

Thus, a POSA would have understood the ʼ639 Patents C18

and C23 as selecting compression code tables on a subregion
basis and encoding “identifiers” for these selected
compression code tables. (EX1002 at ¶202.)

To the extent limitation [10B] of the ʼ583 Patent is not seen

to be an obvious variant of Claim 23 of the ʼ639 Patent, a
POSA would have understood H.263+ as teaching limitation
[10B], including selecting compression code tables on a
subregion basis and encoding identifiers for these selected
compression code tables. (See supra Sections VI.B.2.iii-iv;
EX1002 at ¶¶143-160.)

Additionally, a POSA would have been motivated to

combine Claim 23 of the ʼ639 Patent with H.263+ based on

119
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

’583 Patent ʼ639 Patent’s Claim 23 (which also includes the

Claim Language limitations of Claim 18)

the motivation to combine presented in Section VII.C.2.i

above. (EX1002 at ¶202; see EX1002 at ¶¶191-198.)

Thus, when considering Claim 23 of the ʼ639 Patent in view

of H.263+, a POSA would have understood that the ʼ583
Patent’s limitation [10B] is an indistinct, obvious variant.
(EX1002 at ¶202.)

[10C] a “[ʼ639 Patent Claim 23]

compression
header including 18. A system for storing, transmitting and viewing images,
a cross reference the system comprising:
of the key code
a data compression station configured to store a plurality of
segments to the
compression code tables for conversion of uncompressed
compression
image data representative of individual picture elements to
algorithms.
lossless compressed image data and to execute a
compression routine in which an image data stream is
converted to a compressed file by dividing into subregions
and each subregion compressing in accordance with a
compression code table selected from the plurality of
compression code tables based upon which compression
code table provides optimal lossless compression of the
subregion;

a data storage device for receiving and storing the

compressed file; and

an image decompression station configured to store the

plurality of compression code tables, to access the
compressed file from the data storage device, and to execute
a decompression routine in which the compression code
tables applied to compress the image data stream are applied
to decompress the compressed file to reconstruct the image
data stream.

23. The system of claim 18, wherein the compression routine

includes encoding of identifiers of the selected compression

120
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

’583 Patent ʼ639 Patent’s Claim 23 (which also includes the

Claim Language limitations of Claim 18)

code tables within the compressed file, and wherein the

decompression routine includes analysis of the identifiers
for selection of the same compression code tables for
decompression of the compressed file.” (EX1011 at
Claims 18, 23.)

While the ʼ639 Patent’s Claim 23 teaches “identifiers” being

used to identify the compression code table used by a
decompressor, the ʼ639 Patent’s Claim 23 does not explicitly
teach that, in addition to the identifier, there is a “cross
reference of the key code segments to the compression
algorithms” included in a compression header.

However, as explained in Section VI.B.2.v above, a POSA

would have understood H.263+ as teaching a “cross
reference of the key code segments to the compression
algorithms” included in a compression header. (EX1002 at
¶203; see also EX1002 at ¶¶161-167.)

Additionally, a POSA would have been motivated to

combine Claim 23 of the ʼ639 Patent with H.263+ in this
way based on the motivation to combine presented in
Section VII.C.2.i above. (EX1002 at ¶203; see EX1002 at
¶¶191-198.)

Thus, when considering Claim 23 of the ʼ639 Patent in view

of H.263+, a POSA would have understood that the ʼ583
Patent’s limitation [1C-1] is an indistinct, obvious variant.
(EX1002 at ¶203.)

Thus, for the reasons provided above, Claim 10 of the ʼ583 Patent is a

patentably indistinct, obvious variant of Claim 23 of the ʼ639 Patent in view of

H.263+. (EX1002 at ¶¶191-204.)

121
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

VIII. CONCLUSION
Based on the foregoing, Unified Patents has demonstrated that there are

substantial new questions of patentability for Claims 1 and 10 of the ʼ583 Patent.

Thus, Claims 1 and 10 should be reexamined, rejected, and cancelled.

Respectfully Submitted,

/Theodoros Konstantakopoulos/
Theodoros Konstantakopoulos (Reg. No. 74,155)
Customer No. 132,593
[email protected]
DESMARAIS LLP
230 Park Avenue
New York, NY 10169
Telephone: 212-351-3400
Facsimile: 212-351-3401

Attorney for Third Party Requester

Unified Patents, LLC

122
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

TABLE OF EXHIBITS

Exhibit Description
EX1001 U.S. Patent 6,795,583
EX1002 Declaration of Dr. Joseph Havlicek
EX1003 Prosecution History of U.S. Patent No. 6,795,583.
ITU-T Recommendation H.263, “Video coding for Low Bit Rate
EX1004
Communication,” Published February 1998.
“H.263+: Video Coding at Low Bit Rates,” Guy Cote et al, IEEE
EX1005
Nov. 1998.
“H.264 and MPEG-4 Video Compression,” Ian E.G. Richardson,
EX1006
Published February 2004.
EX1007 U.S. Patent 5,845,088
EX1008 U.S. Patent 5,764,374
EX1009 U.S. Patent 5,748,904
EX1010 U.S. Patent 6,792,151
EX1011 U.S. Patent 7,050,639
“ITU-T Standards Approval,” available at
EX1012 https://www.itu.int/en/ITU-T/about/Pages/approval.aspx (last
accessed April 23rd, 2021)
“ITU-T In Brief,” available at https://www.itu.int/en/ITUT/
EX1013
about/Pages/default.aspx (last accessed April 23rd, 2021)
“ITU-T Standards Development,” available at
EX1014 https://www.itu.int/en/ITU-T/about/Pages/development.aspx (last
accessed April 23rd, 2021)

ATTACHMENTS

i
Request For Ex Parte Reexamination Of U.S. Patent No. 6,795,583
Attorney Docket No. 44382-0007

Attachment No. Description

Attachment 1 Information Disclosure Statement

Attachment 2 Certificate of Service

ii