|
Preface |
5 |
|
|
Organization |
6 |
|
|
Table of Contents |
12 |
|
|
Invited Papers |
12 |
|
|
Virtual Reality: A Knowledge Tool for Cultural Heritage |
15 |
|
|
Introduction |
15 |
|
|
VR and Cultural Heritage Documentation |
16 |
|
|
Suchilquitongo |
17 |
|
|
Digital Reconstruction |
17 |
|
|
Cacaxtla |
17 |
|
|
Techniques Employed |
18 |
|
|
The Registry |
18 |
|
|
The Making of the 3D Models |
20 |
|
|
Describing IXTLI |
21 |
|
|
Results |
22 |
|
|
Further Works |
23 |
|
|
Conclusions |
24 |
|
|
References |
24 |
|
|
Computer Graphics Theory and Applications |
12 |
|
|
Part I: Geometry and Modeling |
12 |
|
|
Using Distance Information for Silhouette Preservation in Mesh Simplification Techniques |
25 |
|
|
Introduction |
25 |
|
|
Previous Work |
26 |
|
|
Mesh Simplification |
26 |
|
|
Digital Distance Transforms |
27 |
|
|
Method Description |
27 |
|
|
View-Dependent Distance Labels Computation |
28 |
|
|
Distance Labels Interpolation for New Points of View |
31 |
|
|
Mesh Simplification |
31 |
|
|
Results |
33 |
|
|
Conclusions and Future Work |
37 |
|
|
References |
38 |
|
|
Closed-Form Solutions for Continuous PCA and Bounding Box Algorithms |
40 |
|
|
Introduction |
40 |
|
|
PCA |
42 |
|
|
Continuous PCA |
43 |
|
|
Evaluation of the Expressions for Continuous PCA |
44 |
|
|
Continuous PCA in $R^2$ |
44 |
|
|
Continuous PCA in $R^3$ |
46 |
|
|
Experimental Results |
48 |
|
|
Evaluation of the PCA and CPCA Bounding Box Algorithms |
49 |
|
|
Evaluation of Other Bounding Box Algorithms |
52 |
|
|
Conclusions |
53 |
|
|
References |
54 |
|
|
Part II: Rendering |
12 |
|
|
An Importance Sampling Method for Arbitrary BRDFs |
55 |
|
|
Introduction |
55 |
|
|
Reflectance Equation and Monte-Carlo Estimation |
56 |
|
|
MC Numerical Estimation of $L_r$ |
56 |
|
|
Sampling the BRDF |
57 |
|
|
Our Algorithm |
59 |
|
|
Building the Adaptive Structures |
59 |
|
|
Obtaining Sample Directions |
60 |
|
|
Quadtree Traversing for Optimal Sampling |
62 |
|
|
Quadtree Set Construction Requirements |
62 |
|
|
Results |
63 |
|
|
Sampling Analytical BRDF Models |
63 |
|
|
Adaptive Sampling of Measured Data |
65 |
|
|
Conclusions |
66 |
|
|
References |
67 |
|
|
Generalization of Single-Center Projections Using Projection Tile Screens |
69 |
|
|
Introduction |
69 |
|
|
Related Work |
71 |
|
|
Real-Time Non-planar Projections |
71 |
|
|
Projection Functions |
73 |
|
|
Adjusting the Orientation of the Projection Camera |
73 |
|
|
Normal-Map Optimization |
74 |
|
|
Concept of Projection Tiles |
75 |
|
|
Tile Features |
76 |
|
|
Projection Tiles and Projection Tile Screens |
76 |
|
|
Mapping Projection-Tile-Screens to Normal-Maps |
77 |
|
|
Implementation Details |
77 |
|
|
Cube Map Creation |
78 |
|
|
Applying Projections |
79 |
|
|
Experimental Results |
79 |
|
|
Application Examples |
79 |
|
|
Problems and Limitations |
81 |
|
|
Conclusions and Outlook |
81 |
|
|
References |
81 |
|
|
Part III: Interactive Environments |
12 |
|
|
Real-Time Generation of Interactive Virtual Human Behaviours |
84 |
|
|
Introduction |
84 |
|
|
Model of Interactive Behaviours |
85 |
|
|
Generating Interactive Behaviours |
86 |
|
|
The Windowed Viterbi Algorithm |
86 |
|
|
Estimating Output Behaviours |
87 |
|
|
Improving the Model for Tracking Accuracy |
89 |
|
|
Results |
90 |
|
|
Assessing the Accuracy of the Generated Behaviour |
91 |
|
|
Visual Evaluation |
93 |
|
|
Assessing the Accuracy of the Tracking Results |
95 |
|
|
Conclusions |
95 |
|
|
References |
96 |
|
|
CoGenIVE: Building 3D Virtual Environments Using a Model Based User Interface Design Approach |
97 |
|
|
Introduction |
97 |
|
|
Related Work |
98 |
|
|
The VR-DeMo Process |
99 |
|
|
Dialog Model |
100 |
|
|
Defining the States |
100 |
|
|
Handling Input |
101 |
|
|
Presentation Model |
102 |
|
|
Interaction Description |
103 |
|
|
NiMMiT Basic Primitives |
104 |
|
|
Additional Features |
104 |
|
|
Tool Support |
105 |
|
|
Application Prototypes |
106 |
|
|
Practical Use of CoGenIVE |
107 |
|
|
Conclusions |
108 |
|
|
References |
108 |
|
|
Computer Vision Theory and Applications |
12 |
|
|
Part I: Image Formation and Processing |
12 |
|
|
Fast Medial Axis Extraction Algorithm on Tubular Large 3D Data by Randomized Erosion |
111 |
|
|
Introduction |
111 |
|
|
Basic Notions |
112 |
|
|
Morphological Operators on Binary Images |
112 |
|
|
Accelerated Thinning and Skeleton Definition |
113 |
|
|
Methods |
114 |
|
|
Skeletonization Method |
114 |
|
|
Algorithm Details |
114 |
|
|
Results |
117 |
|
|
Fast Morphologic Operators on Binary Images |
117 |
|
|
Fast Skeletonization Algorithm |
117 |
|
|
Conclusions |
121 |
|
|
References |
122 |
|
|
Self-calibration of Central Cameras from Point Correspondences by Minimizing Angular Error |
123 |
|
|
Introduction |
123 |
|
|
Central Camera Models |
124 |
|
|
Image Formation in Central Cameras |
124 |
|
|
Radial Projection Models |
126 |
|
|
Backward Models |
126 |
|
|
Self-calibration Method |
127 |
|
|
Minimization of Angular Error for Two Views |
127 |
|
|
Constraints on Camera Parameters |
129 |
|
|
Robustness for Outliers |
129 |
|
|
Three Views |
129 |
|
|
Experiments |
130 |
|
|
Synthetic Data |
130 |
|
|
Real Data |
133 |
|
|
Discussion |
134 |
|
|
Conclusions |
135 |
|
|
References |
135 |
|
|
Image Filtering Based on Locally Estimated Geodesic Functions |
137 |
|
|
Introduction |
137 |
|
|
Similarity Measure Based on Geodesic Time |
139 |
|
|
Geodesic Time on Greylevel Images |
139 |
|
|
A New Geodesic Similarity Measure |
140 |
|
|
Geodesic $\Sigma$-Filter Depending on Image Gradient |
140 |
|
|
Estimation of the Similarity Measure |
140 |
|
|
Design of the Geodesic Filter |
141 |
|
|
Dealing with Multispectral Images |
142 |
|
|
Geodesic $\Delta$-Filter Accounting for Image Variations |
143 |
|
|
Experiments |
144 |
|
|
Implementation |
144 |
|
|
Results, Evaluation and Comparison with Other Methods |
144 |
|
|
Limitations and Improvements |
145 |
|
|
Conclusions |
146 |
|
|
References |
147 |
|
|
Part II: Image Analysis |
13 |
|
|
Computation of Left Ventricular Motion Patterns Using a Normalized Parametric Domain |
149 |
|
|
Introduction |
149 |
|
|
Left Ventricular Function Estimation |
151 |
|
|
Normalized Parametric Domain |
154 |
|
|
Initial Surface Fitting |
154 |
|
|
General Surface Fitting |
155 |
|
|
Regional Analysis of the LV |
156 |
|
|
Results |
157 |
|
|
Mean Motion Patterns |
159 |
|
|
Correlation between Motion and HVMB |
159 |
|
|
Conclusions and Future Work |
160 |
|
|
References |
160 |
|
|
Improving Geodesic Invariant Descriptors through Color Information |
162 |
|
|
Introduction |
162 |
|
|
Related Works |
163 |
|
|
Coloring Geodesic Invariant Features |
164 |
|
|
Fast Marching Algorithm in RGB Space |
165 |
|
|
Building the Geodesic Color Descriptor |
166 |
|
|
Color Invariants Selection |
167 |
|
|
Experimental Results |
170 |
|
|
Discussion and Future Works |
174 |
|
|
References |
174 |
|
|
On Head Pose Estimation in Face Recognition |
176 |
|
|
Introduction |
176 |
|
|
Feature Extraction |
177 |
|
|
Local Energy Model |
177 |
|
|
Proposed Feature Description |
178 |
|
|
Pose Estimation |
180 |
|
|
PIE Database |
180 |
|
|
Proposed Approach |
181 |
|
|
Experimental Setup and Results |
185 |
|
|
Test Results for Seen Imaging Conditions |
185 |
|
|
Test Results for Previously Unseen Illumination Conditions |
186 |
|
|
Test Results for Unseen Poses |
187 |
|
|
Conclusions and Discussion |
187 |
|
|
References |
188 |
|
|
Part III: Image Understanding |
13 |
|
|
Edge-Based Template Matching with a Harmonic Deformation Model |
190 |
|
|
Introduction |
190 |
|
|
Related Work |
191 |
|
|
Main Contributions |
192 |
|
|
Deformable Shape-Based Matching |
192 |
|
|
Shape Model Generation |
192 |
|
|
Deformable Metric Based on Local Edge Patches |
193 |
|
|
Deformable Shape Matching |
195 |
|
|
Harmonic Deformation Model |
196 |
|
|
Experiments |
197 |
|
|
Comparison with Descriptor-Based Matching |
197 |
|
|
Simulated TPS and Harmonic Deformation |
198 |
|
|
Real World Experiments |
199 |
|
|
Conclusions |
200 |
|
|
References |
200 |
|
|
Implementation of a Model for Perceptual Completion in $R^2 × S^1$ |
202 |
|
|
Introduction |
202 |
|
|
Theoretical Background |
204 |
|
|
Lifting of the Image Level Lines in a 3D Space |
204 |
|
|
The Tangent Bundle and the Integral Curves |
204 |
|
|
Curve Length's and Metric of the Space |
205 |
|
|
The Lifted Surface as an Implicit Function |
206 |
|
|
Sub-riemannian Differential Operators |
207 |
|
|
Differential Geometry of the Surface |
208 |
|
|
The Completion Model |
208 |
|
|
Basic Model |
208 |
|
|
Algorithmic Implementation |
209 |
|
|
Multiple Concentration |
210 |
|
|
Numerical Scheme |
211 |
|
|
Experiments and Results |
212 |
|
|
Macula Cieca Example |
212 |
|
|
Occlusion Example |
213 |
|
|
Conclusions |
214 |
|
|
References |
214 |
|
|
Data Compression - A Generic Principle of Pattern Recognition? |
216 |
|
|
Introduction |
216 |
|
|
Compression for Recognition |
218 |
|
|
Method |
218 |
|
|
Theoretical Background |
218 |
|
|
Compression Algorithms |
219 |
|
|
Discussion of the Similarity Measure |
219 |
|
|
Experiments |
219 |
|
|
Object Recognition |
219 |
|
|
Texture Classification |
224 |
|
|
Image Retrieval |
224 |
|
|
Conclusions |
224 |
|
|
References |
225 |
|
|
Hierarchical Evaluation Model: Extended Analysis for 3D Face Recognition |
227 |
|
|
Introduction |
227 |
|
|
Related Works |
228 |
|
|
3D Face Matching |
229 |
|
|
The Surface Interpenetration Measure |
229 |
|
|
SA-Based Approach for Range Image Registration |
230 |
|
|
Modified SA-Based Approach for Range Image Registration |
231 |
|
|
3D Face Authentication |
232 |
|
|
Hierarchical Evaluation Model |
233 |
|
|
Experimental Results |
234 |
|
|
Alignment Results |
234 |
|
|
Analysis of the Hierarchical Evaluation Model |
234 |
|
|
Final Remarks |
236 |
|
|
References |
237 |
|
|
Estimation of 3D Instantaneous Motion of a Ball from a Single Motion-Blurred Image |
239 |
|
|
Introduction |
239 |
|
|
Related Works |
240 |
|
|
Problem Formulation |
241 |
|
|
Blurred Image Formation |
241 |
|
|
Blur on the Ball Surface |
241 |
|
|
Image Analysis |
242 |
|
|
Alpha Matting |
242 |
|
|
Blur Analysis |
243 |
|
|
Reconstruction Technique |
243 |
|
|
Null Translation |
244 |
|
|
Recovering the Ball 3D Position and Velocity |
245 |
|
|
Recovering Spin in the General Case |
246 |
|
|
Experiments |
247 |
|
|
Discussion and Conclusions |
250 |
|
|
References |
250 |
|
|
Part IV: Motion, Tracking and Stereo Vision |
13 |
|
|
Integrating Current Weather Effectsin to Urban Visualization |
252 |
|
|
Introduction |
252 |
|
|
Related Work |
253 |
|
|
Data Retrieval and Analysis |
254 |
|
|
Digital 3D City Model |
255 |
|
|
Steerable Weather-Camera |
255 |
|
|
Weather Radar and Web-Based Weather Services |
257 |
|
|
Rendering Techniques for Certain Weather Effects |
259 |
|
|
Atmospheric Rendering |
259 |
|
|
Rendering of Clouds |
259 |
|
|
Rendering of Rain and Snow |
261 |
|
|
Rendering of Fog |
261 |
|
|
Subjective Evaluation |
261 |
|
|
Tasks |
261 |
|
|
Results |
262 |
|
|
Conclusions |
263 |
|
|
References |
264 |
|
|
Guided KLT Tracking Using Camera Parameters in Consideration of Uncertainty |
266 |
|
|
Introduction |
266 |
|
|
Problem Statement and Motivation |
266 |
|
|
Literature Review |
267 |
|
|
KLT Tracking |
267 |
|
|
Using Intrinsic and Extrinsic Camera Parameters |
268 |
|
|
In Consideration of Uncertainty |
269 |
|
|
Experimental Results |
270 |
|
|
Trail Length Evaluation |
270 |
|
|
Accuracy Evaluation |
272 |
|
|
Conclusions and Outlook |
274 |
|
|
References |
275 |
|
|
Automated Object Identification and Position Estimation for Airport Lighting Quality Assessment |
276 |
|
|
Introduction |
276 |
|
|
Model-Based (MB) Tracking |
278 |
|
|
Camera Positioning |
280 |
|
|
Distortion Correction |
281 |
|
|
Multi Frame-Based Estimation |
282 |
|
|
Constraints |
283 |
|
|
Position and Orientation Results |
283 |
|
|
Luminous Intensity Estimation for Quality Assessment |
286 |
|
|
Luminous Intensity Estimation Results |
286 |
|
|
Concluding Remarks |
288 |
|
|
References |
289 |
|
|
Author Index |
290 |
|