Unifying Perspectives in Computational and Robot Vision

Front Cover
Danica Kragic, Ville Kyrki
Springer Science & Business Media, Jun 6, 2008 - Computers - 206 pages

Currently there is a gap between the research conducted in computer vision and robotics communities. There are many characteristics in common in computer vision research and vision research in robotics. Despite having these common interests, however, “pure” computer vision has seen significant theoretical and methodological advances during the last decade which many of the robotics researchers are not fully aware of. On the other hand, the manipulation and control capabilities of robots as well as the range of application areas have developed greatly. In robotics, vision can not be considered an isolated component, but it is instead a part of a system resulting in an action. Thus, in robotics the vision research must include consideration of the control of the system, in other words, the entire perception-action loop. This requires that a holistic system approach is useful and could provide significant advances in this application domain. Assembled here is a collection of some of the state of the art methods that are using computer vision and machine learning techniques and apply them in robotic applications.

 

What people are saying - Write a review

We haven't found any reviews in the usual places.

Contents

Recent Trends in Computational and Robot Vision
1
12 Perception and Action
2
13 Mapping the Environment SLAM and vSLAM
3
14 From Maps to Understanding the Environment
5
15 Future Directions
7
References
8
Extracting Planar Kinematic Models Using Interactive Perception
11
22 Related Work
13
642 Modeling the Fiducials
94
643 Registration as a Pose Estimation Problem
96
644 Experimental Validation
97
65 Conclusion
98
References
99
A Sliding Window Filter for Incremental SLAM
103
72 Nonlinear Least Squares SLAM
104
722 Kinematic Process Model
105

23 Interactive Perception
14
24 Obtaining Kinematic Models Through Forceful Interactions
15
242 Tracking Objects
16
244 Graph Analysis
17
25 Experimental Results
18
26 Conclusion
21
References
22
People Detection Using Multiple Sensors on a Mobile Robot
25
32 Related Work
26
33 Partbased Model
28
332 Missing Detections and Clutter
29
334 Learning Model Parameters
30
335 Detection
31
342 Extending the Partbased Model
32
35 Experiments
33
352 Multiple Sensor People Detection
34
353 Recognition from the Robot
36
36 Conclusions
38
References
39
Perceiving Objects and Movements to Generate Actions on a Humanoid Robot
41
42 Active Humanoid Head
42
421 System Requirements
43
422 Head Motor System
44
43 Perceiving Objects and Movements
45
432 Object Representations for Actions
46
44 Action Representation
49
45 Imitation on Objects
50
451 Adaptation of Movements to the Given Situation
51
452 Generalization Across Movements
52
46 Discussion and Conclusions
54
Walds Sequential Analysis for Timeconstrained Vision Problems
57
52 The Twoclass Sequential Decisionmaking Problem
58
521 Sequential Probability Ratio Test
59
53 WaldBoost
61
532 Likelihood Ratio Estimation with AdaBoost
62
533 The WaldBoost Algorithm
63
535 Classification
64
536 Algorithm Details
65
541 Experiments
66
55 WaldBoost Trained Fast Interest Region Detection
68
551 HessianLaplace WaldBoost Classifier
69
552 Experiments
70
56 Robust Estimation of Model Parameters RANSAC with Optimal Sequential Verification
71
561 The Optimal Sequential Test
73
562 Estimation of δ and ε
75
57 Conclusions
76
Pose Estimation and Feature Tracking for Robot Assisted Surgery with Medical Imaging
79
62 Pose Estimation of a Laparoscopic Instrument with Landmarks
81
622 Pose Estimation with Multiple Features
85
63 Pose Estimation of a Laparoscopic Instrument without Landmarks
88
632 Direct Pose Computation
90
64 Pose Estimation of Stereotactic Landmarks
92
724 Point Estimation
106
73 Sparsity in the System Equations
107
74 The Sliding Window Filter
109
75 Conclusions
110
References
112
Topological and Metric Robot Localization through Computer Vision Techniques
113
82 Visionbased Hierarchical Localization
115
822 Structure From Motion SFM Metric Localization
118
83 Local Image Features
121
84 Experiments
122
842 Metric Localization
124
85 Conclusion
125
References
127
More Vision for SLAM
129
92 Overview of the SLAM
130
922 Vision and SLAM
131
93 Vision and Mapping in SLAM
132
932 Loop Closing
142
94 Discussion
143
References
145
Maps Objects and Contexts for Robots
149
102 Structuring Space
152
103 Featurebased Object Discovery
153
104 Visual Context
156
105 Acting in Space
158
106 Summary
159
References
160
VisionBased Navigation Strategies
163
1111 Related Work
165
112 Navigation Alternatives
166
1121 Mapbased Navigation
167
1122 ImageBased Navigation
171
113 Monocular VSLAM Approach
175
1132 Motion Recovery
177
1133 Open Challenges
179
114 Results
182
1142 Reconstruction Results
183
115 Conclusions
184
ImageBased Visual Servoing with Extra Task Related Constraints in a General Framework for SensorBased Robot Systems
187
122 Application
189
1232 Object and Feature Frames
191
1233 Feature Coordinates
192
1234 Task Specification
194
1241 Definition of the Constraints
196
1242 Obtaining the Constraint Matrices
197
1244 Obtaining the Weighting Matrix
198
125 Results
199
126 Conclusion
201
References
202
Index
205
Copyright

Other editions - View all

Common terms and phrases

Popular passages

Page 185 - Dementhon and LS Davis. Model-based object pose in 25 lines of code.