0.25
0.5
0.75
1.25
1.5
1.75
2
Rich Probabilistic Models for Holistic Scene Understanding
Published on Aug 23, 201111138 Views
Research focuses on using probabilistic models and machine learning to understand complex domains that involve large amounts of uncertainty.
Related categories
Chapter list
Rich Probabilistic Models for Holistic Scene Understanding00:00
A Tale of Three Bridges00:11
From Perception to Understanding00:54
Object Detection02:19
Basic Object Detection03:23
Outline - 103:54
Outline - 204:20
Indoor Scene Reconstruction04:29
Learning with Clutter05:12
Energy Function06:42
Latent Variables are Tricky07:57
Grounding Latent Variables08:39
Effect of Informed Prior09:40
Experimental Results10:02
Comparison to labeled clutter - 111:06
Comparison to labeled clutter - 211:27
Outline - 311:46
Scene Segmentation11:59
Region-Based Model - 112:26
Region-Based Model - 213:16
Example Results14:24
Application: 3d Reconstruction14:57
Example 3D Reconstructions15:59
Object Detection16:20
Examples17:58
Detection Performance18:45
Latent Variables Revisited - 119:35
Latent Variables Revisited - 220:34
Real Multi-Class Segmentation21:06
“Fully” Supervised Data - 121:40
“Fully” Supervised Data - 222:03
Weakly Supervised Data - 122:27
Weakly Supervised Data - 222:48
Diverse Data23:05
Latent Variable Formulation - 123:14
Latent Variable Formulation - 224:10
Latent Variable Formulation - 324:48
Learning with Diverse Data25:16
Outline - 426:09
Max-Margin Training26:31
Structured Max-Margin Training28:02
Max-Margin Structured Prediction29:56
Latent SVM31:24
CCCP32:31
Easy - 133:25
Easy - 233:46
Hard - 133:58
Hard - 234:26
Inspiration: Human Learning - 134:42
Inspiration: Human Learning - 235:06
Curriculum Learning35:16
Self-Paced Learning - 135:49
Self-Paced Learning - 236:24
Self-Paced Learning - 337:38
Self-Paced Learning - 438:00
Simple Example: Object Detection38:33
Imputation – Iteration 1 - 139:16
Imputation – Iteration 539:53
Imputation – Iteration 940:03
Imputation – Iteration 1340:12
Self-Paced Learning - 540:19
Self-Paced Learning - 641:11
Image Segmentation Revisited41:32
Outline - 642:28
Model Selection42:39
Human learning revisited - 142:55
Human learning revisited - 243:16
Human learning revisited - 343:18
Human learning revisited - 443:22
Human learning revisited - 543:23
Human learning revisited - 643:26
Multiple Kernel Learning - 143:31
Multiple Kernel Learning - 244:13
Self-Paced Multiple Kernel Learning - 144:56
Self-Paced Multiple Kernel Learning245:25
SPMKL Behavior - 145:47
SPMKL Behavior - 246:27
Imputation – Iteration 1 - 246:53
Imputation – Iteration 347:12
Imputation – Iteration 647:31
Imputation – Iteration 1047:37
Imputation – At Convergence47:42
Classification Accuracy47:52
Bounding Box Imputation48:36
DNA Binding Motif49:03
Conclusion I49:17
Conclusion II50:03
Conclusion III50:35
The Future of Education50:55
The Future of Machine Learning?51:22
Acknowledgments51:38