0% found this document useful (0 votes)
19 views40 pages

Computer Vision - 01 Introduction

The document outlines a course on Computer Vision, detailing grading criteria, textbooks, and essential topics such as image processing, feature detection, segmentation, and object recognition. It covers various algorithms and applications in the field, including optical character recognition, medical imaging, and automotive safety. The course aims to provide a comprehensive understanding of both theoretical and practical aspects of computer vision.
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
0% found this document useful (0 votes)
19 views40 pages

Computer Vision - 01 Introduction

The document outlines a course on Computer Vision, detailing grading criteria, textbooks, and essential topics such as image processing, feature detection, segmentation, and object recognition. It covers various algorithms and applications in the field, including optical character recognition, medical imaging, and automotive safety. The course aims to provide a comprehensive understanding of both theoretical and practical aspects of computer vision.
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/ 40

Computer Vision

(240069)

Introduction

Muhammad Tariq Mahmood


[email protected]
School of Computer Science and Engineering
Korea University of Technology and Education
1
Computer Vision (240069)
 Grading
 2 Assignments/Presentations (40%)
 Term Project (paper + presentation) (30%)
 Final exam (30%)

 Textbook
 Computer Vision: Algorithms and Applications by R. Szeliski
(http://szeliski.org/Book/)
 Additional References
 Computer vision: models, learning and inference by Simon J.D. Prince
(http://www.computervisionmodels.com/)
 Computer Vision: A Modern Approach by D. Forsyth and J. Ponce
 Multiple View Geometry by R. Hartly and A. Zisserman

2
What is Computer Vision?
 Vision
 It is the process of discovering what is present in the world and where
it is by looking.

 Computer Vision
 It is the study of analysis of pictures and videos in order to achieve
results similar to those as by men.

3
What is Computer Vision?

Pattern Machine
Recognition Learning

Mathematics
Image Computer & Probability
Processing Vision

Computer Optics
Graphics

4
Contents

Computer Vision Algorithms and Applications by Richard Szeliski Springer


5
Image Formation
 Geometric primitives
 2D transformations
 3D transformations
 3D rotations
 3D to 2D projections
 Lens distortions
 Photometric image formation
 Lighting
 Reflectance and shading
 Optics
 The digital camera
 Sampling and aliasing
 Color
 Compression

6
Image Processing
 Point operators
 Pixel transforms
 Color transforms
 Compositing and matting
 Histogram equalization
 Application: Tonal adjustment
 Linear filtering
 Separable filtering
 Examples of linear filtering
 Band-pass and steerable filters
 More neighborhood operators
 Non-linear filtering
 Morphology
 Distance transforms
 Connected components

7
Feature Detection and Matching
 Points and patches
 Feature detectors
 Feature descriptors
 Feature matching
 Feature tracking
 Application: Performance-driven animation
 Edges
 Edge detection
 Edge linking
 Application: Edge editing and enhancement
 Lines
 Successive approximation
 Hough transforms
 Vanishing points
 Application: Rectangle detection

8
Segmentation
 Active contours
 Snakes
 Dynamic snakes and
CONDENSATION
 Scissors
 Level Sets
 Application: Contour tracking and
rotoscoping
 Split and merge
 Watershed
 Region splitting (divisive clustering)
 Region merging (agglomerative
clustering)
 Graph-based segmentation
 Probabilistic aggregation
 Mean shift and mode finding
 K-means and mixtures of Gaussians
 Mean shift
 Normalized cuts
 Graph cuts and energy-based methods
 Application: Medical image
segmentation
9
Segmentation
 Active contours
 Snakes
 Dynamic snakes and
CONDENSATION
 Scissors
 Level Sets
 Application: Contour tracking and
rotoscoping
 Split and merge
 Watershed
 Region splitting (divisive clustering)
 Region merging (agglomerative
clustering)
 Graph-based segmentation
 Probabilistic aggregation
 Mean shift and mode finding
 K-means and mixtures of Gaussians
 Mean shift
 Normalized cuts
 Graph cuts and energy-based methods
 Application: Medical image
segmentation
10
Feature-based alignment
 2D and 3D feature-based alignment
 2D alignment using least squares
 Application: Panography
 Iterative algorithms
 Robust least squares and RANSAC
 3D alignment
 Pose estimation
 Linear algorithms
 Iterative algorithms
 Application: Augmented reality
 Geometric intrinsic calibration
 Calibration patterns
 Vanishing points
 Application: Single view metrology
 Rotational motion
 Radial distortion

11
Structure from motion
 Triangulation
 Two-frame structure from motion
 Projective (uncalibrated)
reconstruction
 Self-calibration
 Application: View morphing
 Factorization
 Perspective and projective
factorization
 Application: Sparse 3D model
extraction
 Bundle adjustment
 Exploiting sparsity
 Application: Match move and
augmented reality
 Uncertainty and ambiguities
 Application: Reconstruction from
Internet photos
 Constrained structure and motion
 Line-based techniques
 Plane-based techniques

12
Dense motion estimation
 Translational alignment
 Hierarchical motion estimation
 Fourier-based alignment
 Incremental refinement
 Parametric motion
 Application: Video stabilization
 Learned motion models
 Spline-based motion
 Application: Medical image
registration
 Optical flow
 Multi-frame motion estimation
 Application: Video denoising
 Application: De-interlacing
 Layered motion
 Application: Frame
interpolation
 Transparent layers and
reflections
13
Image Stitching
 Motion models
 Planar perspective motion
 Application: Whiteboard and
document scanning
 Rotational panoramas
 Gap closing
 Application: Video summarization
and compression
 Cylindrical and spherical coordinates
 Global alignment
 Bundle adjustment
 Parallax removal
 Recognizing panoramas
 Direct vs. feature-based alignment
 Compositing
 Choosing a compositing surface
 Pixel selection and weighting (de-
ghosting)
 Application: Photomontage
 Blending

14
Computational Photography
 Photometric calibration
 Radiometric response function
 Noise level estimation
 Vignetting
 Optical blur (spatial response)
estimation
 High dynamic range imaging
 Tone mapping
 Application: Flash photography
 Super-resolution and blur removal
 Color image demosaicing
 Application: Colorization
 Image matting and compositing
 Blue screen matting
 Natural image matting
 Optimization-based matting
 Smoke, shadow, and flash matting
 Video matting
 Texture analysis and synthesis
 Application: Hole filling and
inpainting
 Application: Non-photorealistic
rendering 15
Stereo correspondence
 Epipolar geometry
 Rectification
 Plane sweep
 Sparse correspondence
 3D curves and profiles
 Dense correspondence
 Similarity measures
 Local methods
 Sub-pixel estimation and uncertainty
 Application: Stereo-based head
tracking
 Global optimization
 Dynamic programming
 Segmentation-based techniques
 Application: Z-keying and
background replacement
 Multi-view stereo
 Volumetric and 3D surface
reconstruction
 Shape from silhouettes

16
3D reconstruction
 Shape from X
 Shape from shading and photometric stereo
 Shape from texture
 Shape from focus
 Active rangefinding
 Range data merging
 Application: Digital heritage
 Surface representations
 Surface interpolation
 Surface simplification
 Geometry images
 Point-based representations
 Volumetric representations
 Implicit surfaces and level sets
 Model-based reconstruction
 Architecture
 Heads and faces
 Application: Facial animation
 Whole body modeling and tracking
 Recovering texture maps and albedos
 Estimating BRDFs
 Application: 3D photography

17
Image-based rendering
 View interpolation
 View-dependent texture maps
 Application: Photo Tourism
 Layered depth images
 Impostors, sprites, and layers
 Light fields and Lumigraphs
 Unstructured Lumigraph
 Surface light fields
 Application: Concentric mosaics
 Environment mattes
 Higher-dimensional light fields
 The modeling to rendering
continuum
 Video-based rendering
 Video-based animation
 Video textures
 Application: Animating pictures
 3D Video
 Application: Video-based
walkthroughs

18
 Object detection
Recognition
 Face detection
 Pedestrian detection
 Face recognition
 Eigenfaces
 Active appearance and 3D shape models
 Application: Personal photo collections
 Instance recognition
 Geometric alignment
 Large databases
 Application: Location recognition
 Category recognition
 Bag of words
 Part-based models
 Recognition with segmentation
 Application: Intelligent photo editing
 Context and scene understanding
 Learning and large image collections
 Application: Image search
 Recognition databases and test sets

19
Computer Vision Levels
 Low-level vision (early vision)
 Image formation
 Edge detection & image filtering
 Optical flow
 Segmentation
 Shape matching
 Stereopsis
 Mid-level vision
 Object tracking
 Human motion analysis
 High-level vision
 Object recognition
 Event detection
 Scene & video understanding
20
Computer Vision Algorithms
 Structure from motion algorithms can reconstruct a sparse 3D
point model of a large complex scene from hundreds of partially
overlapping photographs(Snavely, Seitz, and Szeliski 2006)c
2006 ACM.

21
Computer Vision Algorithms
 Stereo matching algorithms can build a detailed 3D model of a
building from hundreds of differently exposed photographs
taken from the Internet (Goesele, Snavely, Curless et al. 2007) c
2007 IEEE.

22
Computer Vision Algorithms
 Person tracking algorithms can track a person walking in front
of a cluttered background (Sidenbladh, Black, and Fleet 2000) c
2000 Springer

23
Computer Vision Algorithms
 Face detection algorithms, coupled with color-based clothing
and hair detection algorithms, can locate and recognize the
individuals in this image (Sivic, Zitnick, and Szeliski 2006) c
2006 Springer.

24
Computer Vision Applications
 Optical character recognition (OCR): reading handwritten
postal codes on letters (Figure 1.4a) and automatic number plate
recognition (ANPR)

25
Computer Vision Applications
 Machine inspection: rapid parts inspection for quality assurance
using stereo vision with specialized illumination to measure
tolerances on aircraft wings or auto body parts or looking for
defects in steel castings using X-ray vision;

26
Computer Vision Applications
 Retail: object recognition for automated checkout

27
Computer Vision Applications
 Medical imaging: registering pre-operative and intra-operative
imagery or performing long-term studies of people’s brain
morphology as they

28
Computer Vision Applications
 Automotive safety: detecting unexpected obstacles such as
pedestrians on the street, under conditions where active vision
techniques such as radar or lidar do not work well

29
Computer Vision Applications
 Surveillance: monitoring for intruders, analyzing highway
traffic, and monitoring pools for drowning victims;

30
Computer Vision Application
 Fingerprint recognition and biometrics: for automatic access
authentication as well

 3D model building (photogrammetry): fully automated


construction of 3D models from aerial photographs

31
Computer Vision Applications

 image stitching: merging different views (Szeliski and Shum


1997) c 1997 ACM

32
Computer Vision Applications

 Exposure bracketing: merging different exposures

33
Computer Vision Applications

 Morphing:blending between two photographs (Gomes, Darsa,


Costa et al. 1999) c 1999 Morgan Kaufmann

34
Computer Vision Applications

 Turning a collection of photographs into a 3D model (Sinha,


Steedly, Szeliski et al. 2008) c 2008 ACM

35
A Brief History

36
A Brief History

37
A Brief History

38
A Brief History

39
A Brief History

40

You might also like