Quan Image-Based Modeling

1. Auflage 2010
ISBN: 978-1-4419-6679-7
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 251 Seiten

'This book guides you in the journey of 3D modeling from the theory with elegant mathematics to applications with beautiful 3D model pictures. Written in a simple, straightforward, and concise manner, readers will learn the state of the art of 3D reconstruction and modeling.' -Professor Takeo Kanade, Carnegie Mellon University The computer vision and graphics communities use different terminologies for the same ideas. This book provides a translation, enabling graphics researchers to apply vision concepts, and vice-versa, independence of chapters allows readers to directly jump into a specific chapter of interest, compared to other texts, gives more succinct treatment overall, and focuses primarily on vision geometry. Image-Based Modeling is for graduate students, researchers, and engineers working in the areas of computer vision, computer graphics, image processing, robotics, virtual reality, and photogrammetry.

Long Quan is a Professor of the Department of Computer Science and Engineering at the Hong Kong University of Science and Technology. He received a Ph.D. degree in Computer Science from France, and has been a CNRS researcher at INRIA. Professor Quan is a Fellow of the IEEE Computer Society.

Autoren/Hrsg.

Quan, Long

Inhaltsverzeichnis

1;Foreword;6
2;Preface;7
3;Acknowledgements;9
4;Notation;10
5;Contents;11
6;Introduction;15
7;Part I Geometry: fundamentals of multi-view geometry;19
7.1;Geometry prerequisite;20
7.1.1;2.1 Introduction;21
7.1.2;2.2 Projective geometry;21
7.1.2.1;2.2.1 The basic concepts;21
7.1.2.2;2.2.2 Projective spaces and transformations;23
7.1.2.3;2.2.3 Affine and Euclidean specialization;29
7.1.3;2.3 Algebraic geometry;34
7.1.3.1;2.3.1 The simple methods;34
7.1.3.2;2.3.2 Ideals, varieties, and Gr¨obner bases;36
7.1.3.3;2.3.3 Solving polynomial equations with Gr¨obner bases;37
7.2;Multi-view geometry;41
7.2.1;3.1 Introduction;42
7.2.2;3.2 The single-view geometry;42
7.2.2.1;3.2.1 What is a camera?;42
7.2.2.2;3.2.2 Where is the camera?;47
7.2.2.3;3.2.3 The DLT calibration;49
7.2.2.4;3.2.4 The three-point pose algorithm;51
7.2.3;3.3 The uncalibrated two-view geometry;54
7.2.3.1;3.3.1 The fundamental matrix;55
7.2.3.2;3.3.2 The seven-point algorithm;57
7.2.3.3;3.3.3 The eight-point linear algorithm;58
7.2.4;3.4 The calibrated two-view geometry;59
7.2.4.1;3.4.1 The essential matrix;59
7.2.4.2;3.4.2 The five-point algorithm;61
7.2.5;3.5 The three-view geometry;65
7.2.5.1;3.5.1 The trifocal tensor;66
7.2.5.2;3.5.2 The six-point algorithm;70
7.2.5.3;3.5.3 The calibrated three views;75
7.2.6;3.6 The N-view geometry;78
7.2.6.1;3.6.1 The multi-linearities;78
7.2.6.2;3.6.2 Auto-calibration;80
7.2.7;3.7 Discussions;84
7.2.8;3.8 Bibliographic notes;84
8;Part II Computation: from pixels to 3D points;86
8.1;Feature point;87
8.1.1;4.1 Introduction;88
8.1.2;4.2 Points of interest;88
8.1.2.1;4.2.1 Tracking features;88
8.1.2.2;4.2.2 Matching corners;90
8.1.2.3;4.2.3 Discussions;91
8.1.3;4.3 Scale invariance;92
8.1.3.1;4.3.1 Invariance and stability;92
8.1.3.2;4.3.2 Scale, blob and Laplacian;92
8.1.3.3;4.3.3 Recognizing SIFT;93
8.1.4;4.4 Bibliographic notes;94
8.2;Structure from Motion;95
8.2.1;5.1 Introduction;96
8.2.1.1;5.1.1 Least squares and bundle adjustment;96
8.2.1.2;5.1.2 Robust statistics and RANSAC;98
8.2.2;5.2 The standard sparse approach;100
8.2.2.1;5.2.1 A sequence of images;102
8.2.2.2;5.2.2 A collection of images;103
8.2.3;5.3 The match propagation;104
8.2.3.1;5.3.1 The best-first match propagation;104
8.2.3.2;5.3.2 The properties of match propagation;107
8.2.3.3;5.3.3 Discussions;111
8.2.4;5.4 The quasi-dense approach;113
8.2.4.1;5.4.1 The quasi-dense resampling;113
8.2.4.2;5.4.2 The quasi-dense SFM;114
8.2.4.3;5.4.3 Results and discussions;121
8.2.5;5.5 Bibliographic notes;127
9;Part III Modeling: from 3D points to objects;129
9.1;Surface modeling;130
9.1.1;6.1 Introduction;131
9.1.2;6.2 Minimal surface functionals;132
9.1.3;6.3 A unified functional;133
9.1.4;6.4 Level-set method;133
9.1.5;6.5 A bounded regularization method;134
9.1.6;6.6 Implementation;136
9.1.7;6.7 Results and discussions;138
9.1.8;6.8 Bibliographic notes;145
9.2;Hair modeling;146
9.2.1;7.1 Introduction;147
9.2.2;7.2 Hair volume determination;148
9.2.3;7.3 Hair fiber recovery;149
9.2.3.1;7.3.1 Visibility determination;149
9.2.3.2;7.3.2 Orientation consistency;150
9.2.3.3;7.3.3 Orientation triangulation;150
9.2.4;7.4 Implementation;151
9.2.5;7.5 Results and discussions;153
9.2.6;7.6 Bibliographic notes;157
9.3;Tree modeling;158
9.3.1;8.1 Introduction;159
9.3.2;8.2 Branche recovery;162
9.3.2.1;8.2.1 Reconstruction of visible branches;162
9.3.2.2;8.2.2 Synthesis of occluded branches;164
9.3.2.3;8.2.3 Interactive editing;166
9.3.3;8.3 Leaf extraction and reconstruction;168
9.3.3.1;8.3.1 Leaf texture segmentation;168
9.3.3.2;8.3.2 Graph-based leaf extraction;171
9.3.3.3;8.3.3 Model-based leaf reconstruction;174
9.3.4;8.4 Results and discussions;176
9.3.5;8.5 Bibliographic notes;183
9.4;Fac¸ade modeling;185
9.4.1;9.1 Introduction;186
9.4.2;9.2 Fac¸ade initialization;188
9.4.2.1;9.2.1 Initial flat rectangle;189
9.4.2.2;9.2.2 Texture composition;189
9.4.2.3;9.2.3 Interactive refinement;191
9.4.3;9.3 Fac¸ade decomposition;192
9.4.3.1;9.3.1 Hidden structure discovery;192
9.4.3.2;9.3.2 Recursive subdivision;193
9.4.3.3;9.3.3 Repetitive pattern representation;194
9.4.3.4;9.3.4 Interactive subdivision refinement;195
9.4.4;9.4 Fac¸ade augmentation;196
9.4.4.1;9.4.1 Depth optimization;196
9.4.4.2;9.4.2 Cost definition;198
9.4.4.3;9.4.3 Interactive depth assignment;198
9.4.5;9.5 Fac¸ade completion;200
9.4.6;9.6 Results and discussions;200
9.4.7;9.7 Bibliographic notes;205
9.5;Building modeling;207
9.5.1;10.1 Introduction;208
9.5.2;10.2 Pre-processing;209
9.5.3;10.3 Building segmentation;211
9.5.3.1;10.3.1 Supervised class recognition;211
9.5.3.2;10.3.2 Multi-view semantic segmentation;213
9.5.4;10.4 Building partition;215
9.5.4.1;10.4.1 Global vertical alignment;216
9.5.4.2;10.4.2 Block separator;216
9.5.4.3;10.4.3 Local horizontal alignment;217
9.5.5;10.5 Fac¸ade modeling;218
9.5.5.1;10.5.1 Inverse orthographic composition;219
9.5.5.2;10.5.2 Structure analysis and regularization;221
9.5.5.3;10.5.3 Repetitive pattern rediscovery;224
9.5.5.4;10.5.4 Boundary regularization;225
9.5.6;10.6 Post-processing;226
9.5.7;10.7 Results and discussions;227
9.5.8;10.8 Bibliographic notes;232
10;List of Algorithms;234
11;List of Figures;235
12;References;243
13;Index;255