Advances in image processing and understanding: A festschrift for T.S. Huang

Contents......Page 6
Developmental Vision Audition Robots and Beyond......Page 8
1.1.1 The traditional manual development paradigm......Page 9
1.1.2 Is human vision system totally genetically predetermined ?......Page 10
1.1.3 The new autonomous development paradigm......Page 11
1.1.4 The developmental approach......Page 12
1.1.5 Comparison of approaches......Page 13
1.1.6 More tractable......Page 14
1.2.1 Mode of operation: AA-learning......Page 15
1.2.3 Sensory vector representation......Page 16
1.2.5 Innate and learned behaviors......Page 17
1.3 The Mapping Engine: IHDR......Page 18
1.3.1 Regression......Page 19
1.3.2 Clustering in both input and output space......Page 20
1.3.3 IHDR procedure......Page 22
1.3.4 Amnesic average......Page 24
1.3.5 Discriminating subspace......Page 25
1.3.6 The probability-based metric......Page 27
1.3.7 The transition among different likelihoods......Page 28
1.3.8 Computational considerations......Page 30
1.4.1 SAIL robot......Page 31
1.4.2 Autonomous navigation......Page 32
1.4.4 Test for the developmental algorithm SAIL-2......Page 35
1.4.5 Speech recognition......Page 38
1.5 Conclusions......Page 41
Acknowledgements......Page 42
Bibliography......Page 43
1.Introduction......Page 46
2.1 PBVD -formulation and properties......Page 48
2.2 PBVD-based facial animation......Page 50
3.2 Model-based tracking using the PBVD model......Page 52
3 3 Coarse-to-fine framework......Page 54
4 Explanation-based motion tracking......Page 55
4.2 Learning expressions/visemes or action units......Page 56
5.1 Face model initialization......Page 58
5.2 PBVD model-based tracking......Page 59
5.3 Explanation-based tracking......Page 61
References......Page 62
1. Introduction......Page 64
2. Motivation and Previous Work......Page 66
3. GOES Cloud Image Sequences......Page 67
4. System Outline......Page 68
5. Local Nonrigid Motion Model......Page 70
5.1. Formulations......Page 71
5.2. The Affine Motion Model and Cloud Fluid dynamics......Page 72
6.1. Minimization Method and Error-of-Fit Function......Page 73
6.2. Initial Guesses and Initial Depth Assumption......Page 74
6.3. Depth Constraints......Page 76
7.1. Smooth Motion Assumption......Page 78
7.2. Fluid Dynamics......Page 79
7.3. Incorporating the Global Constraints......Page 80
8 Experimental Results......Page 82
9. Validations......Page 84
10. Conclusions and Future Work......Page 87
References......Page 89
1. Introduction......Page 96
2. System Framework......Page 98
3.1. Preprocessing......Page 100
3.2. Map Image Thinning......Page 101
4. Graph Representations of Maps......Page 102
4.1. Graph Conversion......Page 103
4.2. Graph Redundancy and Redundancy Elimination......Page 104
4.3. Super Graph Description......Page 107
5. Map Graph Segmentation and Re-linking......Page 109
6. Rolling Ball: Road Structure Vectorization......Page 113
6.1. Rolling ball method......Page 114
6.2. Road Inter-Junction Detection......Page 116
6.3. Road Network Exploration......Page 118
7.1. Recognize House Structures......Page 119
7.2. House Data Vectorization......Page 120
8. Discussion......Page 122
Reference......Page 130
1. Introduction......Page 132
2.1. Feature Selection......Page 134
2.2. Dynamic Feature Sets......Page 139
3. Discussion......Page 141
3.1. Major Challenges......Page 142
3.2. Capability for Learning......Page 143
References......Page 144
Introduction to Human Facial Feature Extraction......Page 148
Overview of Automated Facial Feature Extraction Algorithm......Page 149
Choosing face template......Page 151
Template consolidation......Page 152
Resolution pyramid of face image......Page 153
Search strategies......Page 154
Face Detection by Coarse-to-fine Multi-resolution Searching......Page 155
Feature Extraction by Global-to-local Matching......Page 156
Feedback Process......Page 157
Experiments......Page 159
Locating face area......Page 160
Matching with facial features......Page 161
Combining features extracted from different face views......Page 164
Results and Conclusions......Page 165
References......Page 168
Learning Based Relevance Feedback in Image Retrieval......Page 170
1. Introduction......Page 171
2. Concepts and Notations......Page 172
3.1. The MARS approach......Page 174
3.2. The MindReader approach......Page 175
4. The Proposed Approach......Page 176
4.1. Problem formulation......Page 177
4.2. Optimal solution for qi......Page 178
4.4. Optimal Solution for u......Page 179
5.1. Data set......Page 180
5.3. Visual features......Page 181
5.4. Performance measures......Page 182
5.6. Results and observations......Page 183
6. Discussions and Conclusions......Page 186
References......Page 188
1 Introduction......Page 190
2.1 Problem formulation......Page 192
2.2 Probability models......Page 193
2.3 Solution......Page 194
2.4 Results......Page 195
3 Parametric Representation of Dense Object Motion Field......Page 196
3.1 Parametric motion of objects......Page 197
3.3 Coding the object boundaries......Page 199
4-2 Spatiotemporal (3-D) Coding of Objects......Page 200
6 Conclusions......Page 201
7 References......Page 202
1. Introduction......Page 206
2. Methodology......Page 208
2.1. Global Motion Estimation and Motion Blob Detection......Page 210
2.2. Texture and Color Analysis......Page 213
2.2.2. Graylevel Co-occurrence Matrix Measures......Page 215
2.2.3. Fractal Dimension Measures......Page 216
2.3. Region Classification and Motion Blob Verification......Page 217
2.4. Shot Summarization......Page 219
2.5. Event Inference......Page 221
3.1. Global Motion Estimation......Page 222
3.3. Region Classification......Page 224
3.4. Shot Summarization......Page 225
3.5.1. Hunt Events......Page 229
3.5.2. Landing Events......Page 231
3.5.3 Rocket Launch Events......Page 234
4. Summary and Discussion......Page 238
References......Page 240
10.1 Introduction......Page 244
10.2.1 Construction of the TRIRF-frame......Page 247
10.2.2 Multi-Mode Coding......Page 249
10.3.1 Basic Considerations......Page 250
10.3.2 Packet Header Specification......Page 252
10.4 Coding Performance Comparisons......Page 253
10.5 Efficiency Analysis of TRIRF Coding......Page 255
10.5.1.1 Problem formulation......Page 258
10.5.1.2 Rate computation for prediction errors......Page 259
10.5.1.4 Rate derivation based on block motion......Page 263
10.5.2 TRIRF Coding: Channels with Error......Page 265
10.5.3 Simulations......Page 273
10.5.3.1 Results for Reliable Channels......Page 274
10.5.3.2 Results with Packet Errors......Page 277
10.6 Conclusions......Page 278
Bibliography......Page 281
Preliminary Comments......Page 284
1 Fundamentals of 2D AM-FM Modeling......Page 285
2 Isolating the Multiple Image Components......Page 290
3 AM-FM Demodulation......Page 292
4 AM-FM Image Segmentation......Page 295
5 AM-FM Reaction-Diffusion for Texture Completion......Page 300
6 Multidimensional Orthogonal FM Transforms......Page 302
References......Page 308
1. Introduction......Page 314
2. The Basic Algorithm With UTTCQ......Page 316
2.1. System Description......Page 317
2.2. Fixed-rate UTTCQ......Page 319
2.3. Bit Allocation Scheme......Page 321
3. Nonuniform Threshold TCQ......Page 323
4. UTTCQ With Block Classification......Page 324
4.1. Description of the Enhanced Scheme......Page 325
4.2. Classification Methods......Page 326
5. Layered Transmission with RCPC Channel Coding......Page 328
5.1. Layered Grouping and RCPC coding......Page 329
6. Experimental Results and Comparisons......Page 331
7. Conclusion and Discussion......Page 333
Acknowledgements......Page 334
References......Page 335
I. INTRODUCTION......Page 338
II. GENERAL PROBLEM AND NOTATION......Page 339
III. 3D-TO-3D CORRESPONDENCES......Page 340
IV. 2D-TO-3D CORRESPONDENCES......Page 341
V. 2D-TO-2D CORRESPONDENCES......Page 344
VI. FUTURE RESEARCH AND OPEN QUESTIONS......Page 350
VII. SUMMARY AND CONCLUSION......Page 351
REFERENCES......Page 352
I. INTRODUCTION......Page 356
II. WHY MULTIPLE MODALITIES IN HCI?......Page 357
III. MODALITIES FOR HCI......Page 359
IV. WHEN TO INTEGRATE THE HCI MODALITIES......Page 361
V. HOW TO INTEGRATE THE HCI MODALITIES......Page 363
VI. MULTIMODAL HCI SYSTEMS AND APPLICATIONS......Page 367
VII. DISCUSSION......Page 369
REFERENCES......Page 370
II. IMAGE ENHANCEMENT......Page 374
III. EFFICIENT PICTURE CODING......Page 381
IV. OPTICAL IMAGE PROCESSING TECHNIQUES......Page 386
V. DIGITAL COMPUTER IMAGE PROCESSING TECHNIQUES......Page 387
VI. ELECTROOPTICAL DEVICES......Page 392
VII. IMAGE DESCRIPTION......Page 393
BIBLIOGRAPHY......Page 394
REFERENCES......Page 395