3D Computer Vision (eBook)

Christian Wöhler received his Diploma in Physics from Würzburg University in 1996, the Doctorate degree in Computer Science from Bonn University in 2000, and the Habilitation in Applied Computer Science from Bielefeld University in 2009. He is working as a research scientist in the Environment Perception department of Daimler Group Research and Advanced Engineering in Ulm. Since 2005 he is a visiting lecturer at the Technical Faculty of Bielefeld University. His scientific interests are in the domain of pattern classification, three-dimensional computer vision, and photogrammetry, with applications in the fields of driver assistance systems, industrial machine vision, and planetary science."

Preface 6
Acknowledgements 9
Contents 10
Part I Methods of 3D Computer Vision 14
Chapter 1 Geometric Approaches to Three-dimensional Scene Reconstruction 15
1.1 The Pinhole Camera Model 15
1.2 Bundle Adjustment Methods 19
1.3 Geometric Aspects of Stereo Image Analysis 21
1.3.1 Euclidean Formulation of Stereo Image Analysis 21
1.3.2 Stereo Image Analysis in Terms of Projective Geometry 24
1.4 Geometric Calibration of Single and Multiple Cameras 29
1.4.1 Methods for Intrinsic Camera Calibration 29
1.4.2 The Direct Linear Transform (DLT) Method 30
1.4.3 The Camera Calibration Method by Tsai (1987) 33
1.4.4 The Camera Calibration Method by Zhang (1999a) 37
1.4.5 The Camera Calibration Method by Bouguet (2007) 39
1.4.6 Self-calibration of Camera Systems from Multiple Views of a Static Scene 40
1.4.7 Semi-automatic Calibration of Multiocular Camera Systems 53
1.4.8 Accurate Localisation of Chequerboard Corners 63
1.5 Stereo Image Analysis in Standard Geometry 74
1.5.1 Image Rectification According to Standard Geometry 74
1.5.2 The Determination of Corresponding Points 78
1.6 Three-dimensional Pose Estimation and Segmentation Methods 99
1.6.1 Pose Estimation of Rigid Objects 100
1.6.2 Pose Estimation of Non-rigid and Articulated Objects 107
1.6.3 Point Cloud Segmentation Approaches 125
Chapter 2 Photometric Approaches to Three-dimensional Scene Reconstruction 139
2.1 Shape from Shadow 139
2.1.1 Extraction of Shadows from Image Pairs 140
2.1.2 Shadow-based Surface Reconstructionfrom Dense Sets of Images 142
2.2 Shape from Shading 144
2.2.1 The Bidirectional Reflectance Distribution Function (BRDF) 144
2.2.2 Determination of Surface Gradients 149
2.2.3 Reconstruction of Height from Gradients 154
2.2.4 Surface Reconstruction Based on Eikonal Equations 156
2.3 Photometric Stereo 158
2.3.1 Classical Photometric Stereo Approaches 159
2.3.2 Photometric Stereo Approaches Based on Ratio Images 160
2.4 Shape from Polarisation 163
2.4.1 Surface Orientation from Dielectric Polarisation Models 163
2.4.2 Determination of Polarimetric Properties of Rough Metallic Surfaces for Three-dimensional Reconstruction Purposes 166
Chapter 3 Real-aperture Approaches to Three-dimensional Scene Reconstruction 170
3.1 Depth from Focus 172
3.2 Depth from Defocus 173
3.2.1 Basic Principles 173
3.2.2 Determination of Small Depth Differences 178
3.2.3 Determination of Absolute Depth Across Broad Ranges 181
Chapter 4 Integrated Frameworks for Three-dimensional Scene Reconstruction 191
4.1 Monocular Three-dimensional Scene Reconstruction at Absolute Scale 192
4.1.1 Combining Motion, Structure, and Defocus 193
4.1.2 Online Version of the Algorithm 194
4.1.3 Experimental Evaluation Based on Tabletop Scenes 195
4.1.4 Discussion 205
4.2 Self-consistent Combination of Shadow and Shading Features 206
4.2.1 Selection of a Shape from Shading Solution Based on Shadow Analysis 207
4.2.2 Accounting for the Detailed Shadow Structure in the Shape from Shading Formalism 210
4.2.3 Initialisation of the Shape from Shading Algorithm Based on Shadow Analysis 212
4.2.4 Experimental Evaluation Based on Synthetic Data 214
4.2.5 Discussion 215
4.3 Shape from Photopolarimetric Reflectance and Depth 216
4.3.1 Shape from Photopolarimetric Reflectance 217
4.3.2 Estimation of the Surface Albedo 221
4.3.3 Integration of Depth Information 222
4.3.4 Experimental Evaluation Based on Synthetic Data 227
4.3.5 Discussion 232
4.4 Stereo Image Analysis of Non-Lambertian Surfaces 233
4.4.1 Iterative Scheme for Disparity Estimation 235
4.4.2 Qualitative Behaviour of the Specular Stereo Algorithm 239
4.5 Three-dimensional Pose Estimation Based on Combinations of Monocular Cues 240
4.5.1 Appearance-based Pose EstimationRelying on Multiple Monocular Cues 241
4.5.2 Contour-based Pose Estimation Using Depth from Defocus 246
Part II Application Scenarios 251
Chapter 5 Applications to Industrial Quality Inspection 252
5.1 Inspection of Rigid Parts 253
5.1.1 Object Detection by Pose Estimation 253
5.1.2 Pose Refinement 257
5.2 Inspection of Non-rigid Parts 262
5.3 Inspection of Metallic Surfaces 265
5.3.1 Inspection Based on Integration of Shadow and Shading Features 265
5.3.2 Inspection of Surfaces with Non-uniform Albedo 266
5.3.3 Inspection Based on SfPR and SfPRD 268
5.3.4 Inspection Based on Specular Stereo 275
5.3.5 Discussion 282
Chapter 6 Applications to Safe Human–Robot Interaction 285
6.1 Vision-based Human-Robot Interaction 285
6.1.1 The Role of Gestures in Human–Robot Interaction 286
6.1.2 Safe Human–Robot Interaction 287
6.1.3 Pose Estimation of Articulated Objects in the Context of Human–Robot Interaction 290
6.2 Object Detection and Tracking in Three-dimensional Point Clouds 299
6.3 Detection and Spatio-temporal Pose Estimation of Human Body Parts 301
6.4 Three-dimensional Tracking of Human Body Parts 304
Chapter 7 Applications to Lunar Remote Sensing 310
7.1 Three-dimensional Surface Reconstruction Methods for Planetary Remote Sensing 311
7.1.1 Topographic Mapping of Solar System Bodies 311
7.1.2 Reflectance Behaviour of Planetary Regolith Surfaces 314
7.2 Three-dimensional Reconstruction of Lunar Impact Craters 318
7.2.1 Shadow-based Measurement of Crater Depth 318
7.2.2 Three-dimensional Reconstruction of Lunar Impact Craters at High Resolution 321
7.3 Three-dimensional Reconstruction of Lunar Wrinkle Ridges and Faults 329
7.4 Three-dimensional Reconstruction of Lunar Domes 332
7.4.1 General Overview of Lunar Mare Domes 332
7.4.2 Observations of Lunar Mare Domes 335
7.4.3 Image-based Determination of Morphometric Data 338
7.4.4 Geophysical Insights Gained from Topographic Data 350
Chapter 8 Conclusion 358
References 365