0.25
0.5
0.75
1.25
1.5
1.75
2
Pseudo-Bound Optimization for Binary Energies
Published on Oct 29, 20142554 Views
High-order and non-submodular pairwise energies are important for image segmentation, surface matching, deconvolution, tracking and other computer vision problems. Minimization of such energies is gen
Related categories
Chapter list
Pseudo-Bound Optimization for Binary Energies00:00
Labeling Problems in Computer Vision00:05
Energy Minimization for Labeling Problem00:25
Basic Pairwise Energies00:38
More difficult energies01:16
Our framework01:53
Example of high-order energy02:26
From model fitting to entropyoptimization03:00
Energy example: color entropy04:03
Pseudo-bound optimization example04:32
One standard approach - 104:48
One standard approach - 205:34
Bound optimization, in general06:34
Local minima examples (for GrabCut)07:09
This work: Pseudo-Bound Optimization - 107:28
This work: Pseudo-Bound Optimization - 207:39
This work: Pseudo-Bound Optimization - 308:04
This work: Pseudo-Bound Optimization - 408:22
This work: Pseudo-Bound Optimization - 508:30
This work: Pseudo-Bound Optimization - 608:36
General form of Pseudo-Bounds - 108:40
General form of Pseudo-Bounds - 208:41
General form of Pseudo-Bounds - 308:45
General form of Pseudo-Bounds - 408:57
General form of Pseudo-Bounds - 509:03
General form of Pseudo-Bounds - 609:20
General form of Pseudo-Bounds - 709:22
General form of Pseudo-Bounds - 809:27
General form of Pseudo-Bounds - 909:35
General form of Pseudo-Bounds - 1009:42
General form of Pseudo-Bounds - 1109:44
Specific pseudo-bounds Example 1 - 109:51
Specific pseudo-bounds Example 1 - 210:01
Specific pseudo-bounds Example 2 - 110:17
Specific pseudo-bounds Example 2 - 210:18
Specific pseudo-bounds Example 2 - 310:21
Specific pseudo-bounds Example 2 - 410:28
Specific pseudo-bounds Example 2 - 510:30
Specific pseudo-bounds Example 3 - 110:39
Specific pseudo-bounds Example 3 - 210:40
Experimental Results11:15
Experiment results (high-order) - 111:19
Experiment results (high-order) - 211:36
Experiment results (high-order) - 311:56
Matching appearance distribution12:09
Experiment results (high-order) - 412:32
Experiment results (pairwise)13:01
Conclusion - 113:34
Conclusion - 213:35
Conclusion - 313:43
Conclusion - 413:51
Conclusion - 513:57
Conclusion - 614:05
Conclusion - 714:12