Scribd
Upload
63 views43 pages

Enjoy Your Failures

Research involves asking questions that can be answered. Questions are formulated based on our current knowledge and experiences, and often aim to confirm existing group opinions. If a prediction is wrong, it provides an opportunity to learn something new. Examples of revising theories based on programming errors or allowing for new variables are given. Achieving perfect color accuracy between digital and paper is challenging due to distortions and other real-world factors. All progress relies on being open to new ideas and willingness to revise theories based on new evidence.

Uploaded by

davidrojasv
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
63 views43 pages

Enjoy Your Failures

Research involves asking questions that can be answered. Questions are formulated based on our current knowledge and experiences, and often aim to confirm existing group opinions. If a prediction is wrong, it provides an opportunity to learn something new. Examples of revising theories based on programming errors or allowing for new variables are given. Achieving perfect color accuracy between digital and paper is challenging due to distortions and other real-world factors. All progress relies on being open to new ideas and willingness to revise theories based on new evidence.

Uploaded by

davidrojasv
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 43

Enjoy your

failures
misfortunes
accidents
What is Research about ?
Research is about Questions

Computers are useless.


They can only give you answers.
Pablo Picasso
( common attribution )
What are good Questions ?

NOAA:
A good scientific question is one that
can have an answer.

Wittgenstein:
The riddle does not exist.
If a question can be put at all,
then it can also be answered.
What does this mean ?

Questions are created and formulated


within our current experience and
knowledge !

Even worse . . .
Humans are social.
We seek approval and group acceptance.
This means:
We ask questions for which
(a) we know the answer and that
(b) reconfirm our group opinion
at least if we are human !

.. the brain releases more of the reward chemical dopamine


when we fall in line with the group consensus
V. Klucharev, Centre for Cognitive Neuroimaging, Nijmegen
(Neuron, vol 61, p 140)
( source: “How to control a herd of humans”, NewScientist 4. Feb. 2009 )
Finding the Line:

If your prediction was correct:


what have you learned ?

If your prediction was wrong:


what might you learn ?
Some (personal) Examples:

Image Enhancement:
A flat histogram has the best
“information”.

but:
„flattening the log()“
looks better ? ? ?
( internal software package )
Conventional wisdom:

Y(i,j) Flat{Y(i,j)} Flat{ log Y(i,j) }


Conventional wisdom:

Image Enhancement:
A flat “log” histogram has the best
appearance.

BUT ! ! !
L*  Y1/3 and not L*  log( Y )
New wisdom:
Flat Log 1/Cube
New wisdom: short-lived !

L´  Y1/3
Programming error !
L´  log( Y1/3 )

What to do, if a programming error


beats your theory ? ? ?
Enjoy !
Some (personal) Examples:

- c(l)d
R (l) = P (l)

How does Colour change


when I change the paper ?
Beer:

- c(l)di 0.07

Ri(l) = Pi(l) 0.06


0.05
0.04
0.03
P2 0.02

R2(l) = R1(l) 0.01

P1 0.00
-0.01
-0.02
was known not to work:

- c(l)d
R (l) = P (l)

Allow variations in surface adhesion !


Beer:

- c(l)di
Ri(l) = Pi(l)

P2 0.07
0.06
R2(l) = R1(l) 0.05
P1 0.04
0.03

d2 0.02

[ ]
P2 d1
0.01
0.00
R2(l) = R1(l) -0.01
P1 -0.02
0.07
0.06

Too bad ! 0.05


0.04
0.03

Ouch ! ! 0.02
0.01
0.00
-0.01
-0.02

0.06
0.07
0.04
0.05
0.02
0.03

0.01 0.00

-0.01 -0.02

-0.03 -0.04

-0.05
-0.06
-0.07
0.07
0.06

Or ? 0.05
0.04
0.03
0.02
0.01
0.00
-0.01
-0.02

0.06
0.07
0.04
0.05
0.02
0.03

0.01 0.00

-0.01 -0.02

-0.03 -0.04

-0.05
-0.06
-0.07
Misfortune:
re-labeled !
The perfect colour

accurate
precise
error free
ΔE = 0

Objective Metrics !
Can we envision:
!%$?§!
Perfect !!
ΔE = 0
Automatic adjustment
Sounds simple
Digital Paper
maybe not . .
Digital Paper

Distortion
likely not . .
Digital Paper

en
wat t dt
schei

Distortion
Scribble
certainly not . .
Digital Paper

en
wat t dt
schei

Distortion
Scribble
Punch Holes
actually: it is hard . .
Digital Paper

en
wat t dt
schei

Distortion
Scribble Coffee Stain
Punch Holes
George Bernard Shaw:

The reasonable man adapts


himself to the world;
the unreasonable one persists in
trying to adapt the world to
himself.
Therefore all progress depends on
the unreasonable man.
Thank
You !!

You might also like