Computer Vision : A Modern Approach

This extraordinary book gives a uniquely modern view of computer vision. Offering a general survey of the whole computer vision enterprise along with sufficient detail for readers to be able to build useful applications, this book is invaluable in providing a strategic overview of computer vision. With extensive use of probabalistic methods—topics have been selected for their importance, both practically and theoretically—the book gives the most coherent possible synthesis of current views, emphasizing techniques that have been successful in building applications. Readers engaged in computer graphics, robotics, image processing, and imaging in general will find this text an informative reference. Key features: \* Application Surveys—Numerous examples, including Image Based Rendering and Digital Libraries \* Boxed Algorithms—Key algorithms broken out and illustrated in pseudo code \* Self-Contained—No need for other references \* Extensive, Detailed Illustrations—Examples of inputs and outputs for current methods \* Programming Assignments—50 programming assignments and 150 exercises ExLib==ave4eva Cover 1 Computer Vision: A Modern Approach 3 © 4 Dedication Page 5 Contents 7 Preface 19 I: IMAGE FORMATION 33 1 Geometric Camera Models 35 1.1 Image Formation 36 1.2 Intrinsic and Extrinsic Parameters 46 1.3 Geometric Camera Calibration 54 1.4 Notes 61 2 Light and Shading 64 2.1 Modelling Pixel Brightness 64 2.2 Inference from Shading 69 2.3 Modelling Interreflection 84 2.4 Shape from One Shaded Image 91 2.5 Notes 93 3 Color 100 3.1 Human Color Perception 100 3.2 The Physics of Color 105 3.3 Representing Color 109 3.4 A Model of Image Color 118 3.5 Inference from Color 122 3.6 Notes 131 II: EARLY VISION: JUST ONE IMAGE 137 4 Linear Filters 139 4.1 Linear Filters and Convolution 139 4.2 Shift Invariant Linear Systems 144 4.3 Spatial Frequency and Fourier Transforms 150 4.4 Sampling and Aliasing 153 4.5 Filters as Templates 163 4.6 Technique: Normalized Correlation and Finding Patterns 164 4.7 Technique: Scale and Image Pyramids 166 4.8 Notes 169 5 Local Image Features 173 5.1 Computing the Image Gradient 173 5.2 Representing the Image Gradient 176 5.3 Finding Corners and Building Neighborhoods 180 5.4 Describing Neighborhoods with SIFT and HOG Features 187 5.5 Computing Local Features in Practice 192 5.6 Notes 192 6 Texture 196 6.1 Local Texture Representations Using Filters 198 6.2 Pooled Texture Representations by Discovering Textons 203 6.3 Synthesizing Textures and Filling Holes in Images 208 6.4 Image Denoising 214 6.5 Shape from Texture 219 6.6 Notes 223 III: EARLY VISION: MULTIPLE IMAGES 227 7 Stereopsis 229 7.1 Binocular Camera Geometry and the Epipolar Constraint 230 7.2 Binocular Reconstruction 233 7.3 Human Stereopsis 235 7.4 Local Methods for Binocular Fusion 237 7.5 Global Methods for Binocular Fusion 242 7.6 Using More Cameras 246 7.7 Application: Robot Navigation 247 7.8 Notes 248 8 Structure from Motion 253 8.1 Internally Calibrated Perspective Cameras 253 8.2 Uncalibrated Weak-Perspective Cameras 262 8.3 Uncalibrated Perspective Cameras 272 8.4 Notes 280 IV: MID-LEVEL VISION 285 9 Segmentation by Clustering 287 9.1 Human Vision: Grouping and Gestalt 288 9.2 Important Applications 293 9.3 Image Segmentation by Clustering Pixels 300 9.4 Segmentation, Clustering, and Graphs 309 9.5 Image Segmentation in Practice 317 9.6 Notes 319 10 Grouping and Model Fitting 322 10.1 The Hough Transform 322 10.2 Fitting Lines and Planes 325 10.3 Fitting Curved Structures 329 10.4 Robustness 331 10.5 Fitting Using Probabilistic Models 338 10.6 Motion Segmentation by Parameter Estimation 345 10.7 Model Selection: WhichModel Is the Best Fit? 351 10.8 Notes 354 11 Tracking 358 11.1 Simple Tracking Strategies 359 11.2 Tracking Using Matching 366 11.3 Tracking Linear Dynamical Models with Kalman Filters 371 11.4 Data Association 381 11.5 Particle Filtering 382 11.6 Notes 394 V: HIGH-LEVEL VISION 397 12 Registration 399 12.1 Registering Rigid Objects 400 12.2 Model-based Vision: Registering Rigid Objects with Projection 407 12.3 Registering Deformable Objects 410 12.4 Notes 420 13 Smooth Surfaces and Their Outlines 423 13.1 Elements of Differential Geometry 425 13.2 Contour Geometry 434 13.3 Visual Events: More Differential Geometry 439 13.4 Notes 449 14 Range Data 454 14.1 Active Range Sensors 454 14.2 Range Data Segmentation 456 14.3 Range Image Registration and Model Acquisition 464 14.4 Object Recognition 470 14.5 Kinect 478 14.6 Notes 485 15 Learning to Classify 489 15.1 Classification, Error, and Loss 489 15.2 Major Classification Strategies 499 15.3 Practical Methods for Building Classifiers 507 15.4 Notes 513 16 Classifying Images 514 16.1 Building Good Image Features 514 16.2 Classifying Images of Single Objects 536 16.3 Image Classification in Practice 544 16.4 Notes 549 17 Detecting Objects in Images 551 17.1 The Sliding Window Method 551 17.2 Detecting Deformable Objects 562 17.3 The State of the Art of Object Detection 567 17.4 Notes 571 18 Topics in Object Recognition 572 18.1 What Should Object Recognition Do? 572 18.2 Feature Questions 576 18.3 Geometric Questions 579 18.4 Semantic Questions 581 VI: APPLICATIONS AND TOPICS 589 19 Image-Based Modeling and Rendering 591 19.1 Visual Hulls 591 19.2 Patch-Based Multi-View Stereopsis 605 19.3 The Light Field 616 19.4 Notes 619 20 Looking at People 622 20.1 HMM’s, Dynamic Programming, and Tree-Structured Models 622 20.2 Parsing People in Images 634 20.3 Tracking People 638 20.4: 3D from2D: Lifting 643 20.5 Activity Recognition 649 20.6 Resources 656 20.7 Notes 658 21 Image Search and Retrieval 659 21.1 The Application Context 659 21.2 Basic Technologies from Information Retrieval 664 21.3 Images as Documents 671 21.4 Predicting Annotations for Pictures 677 21.5 The State of the Art of Word Prediction 686 21.6 Notes 691 VII: BACKGROUND MATERIAL 693 22 Optimization Techniques 695 22.1 Linear Least-Squares Methods 695 22.2 Nonlinear Least-Squares Methods 701 22.3 Sparse Coding and Dictionary Learning 704 22.4 Min-Cut/Max-Flow Problems and Combinatorial Optimization 707 22.5 Notes 714 Bibliography 716 Index 769 A 769 B 770 C 770 D 774 E 776 F 776 G 777 H 778 I 779 J 780 K 780 L 780 M 781 N 783 O 783 P 784 Q 786 R 786 S 787 T 789 U 790 V 790 W 791 Y 791 Z 791 List of Algorithms 792 Спизжено,у,http://avaxhome.ws/blogs/exlib/ Спизжено у http://avaxhome.ws/blogs/exlib/ This is the eBook of the printed book and may not include any media, website access codes, or print supplements that may come packaged with the bound book. Computer Vision: A Modern Approach, 2e, is appropriate for upper-division undergraduate- and graduate-level courses in computer vision found in departments of Computer Science, Computer Engineering and Electrical Engineering. This textbook provides the most complete treatment of modern computer vision methods by two of the leading authorities in the field. This accessible presentation gives both a general view of the entire computer vision enterprise and also offers sufficient detail for students to be able to build useful applications. Students will learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods

Computer Vision: A Modern Approach, 2e, is appropriate for upper-division undergraduate- and graduate-level courses in computer vision found in departments of Computer Science, Computer Engineering and Electrical Engineering.

This textbook provides the most complete treatment of modern computer vision methods by two of the leading authorities in the field. This accessible presentation gives both a general view of the entire computer vision enterprise and also offers sufficient detail for students to be able to build useful applications. Students will learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods