22nd International Joint Conference on Artificial Intelligence (IJCAI), Barcelona 2011

Rich Probabilistic Models for Holistic Scene Understanding

author: Daphne Koller, Stanford University
published: Aug. 23, 2011, recorded: July 2011, views: 993

Slides

Slides
0:00	Rich Probabilistic Models for Holistic Scene Understanding
0:11	A Tale of Three Bridges
0:54	From Perception to Understanding
2:19	Object Detection
3:23	Basic Object Detection
3:54	Outline - 1
4:20	Outline - 2
4:29	Indoor Scene Reconstruction
5:12	Learning with Clutter
6:42	Energy Function
7:57	Latent Variables are Tricky
8:39	Grounding Latent Variables
9:40	Effect of Informed Prior
10:02	Experimental Results
11:06	Comparison to labeled clutter - 1
11:27	Comparison to labeled clutter - 2
11:46	Outline - 3
11:59	Scene Segmentation
12:26	Region-Based Model - 1
12:27	Scene Segmentation
12:30	Region-Based Model - 1
13:16	Region-Based Model - 2
14:24	Example Results
14:57	Application: 3d Reconstruction
15:59	Example 3D Reconstructions
16:20	Object Detection
17:58	Examples
18:45	Detection Performance
19:35	Latent Variables Revisited - 1
20:34	Latent Variables Revisited - 2
21:06	Real Multi-Class Segmentation
21:40	“Fully” Supervised Data - 1
22:03	“Fully” Supervised Data - 2
22:27	Weakly Supervised Data - 1
22:48	Weakly Supervised Data - 2
23:05	Diverse Data
23:14	Latent Variable Formulation - 1
24:10	Latent Variable Formulation - 2
24:48	Latent Variable Formulation - 3
25:16	Learning with Diverse Data
26:09	Outline - 4
26:31	Max-Margin Training
28:02	Structured Max-Margin Training
29:56	Max-Margin Structured Prediction
31:24	Latent SVM
32:31	CCCP
33:25	Easy - 1
33:46	Easy - 2
33:58	Hard - 1
34:26	Hard - 2
34:42	Inspiration: Human Learning - 1
35:06	Inspiration: Human Learning - 2
35:16	Curriculum Learning
35:49	Self-Paced Learning - 1
36:24	Self-Paced Learning - 2
37:38	Self-Paced Learning - 3
38:00	Self-Paced Learning - 4
38:33	Simple Example: Object Detection
39:16	Imputation – Iteration 1 - 1
39:53	Imputation – Iteration 5
40:03	Imputation – Iteration 9
40:12	Imputation – Iteration 13
40:19	Self-Paced Learning - 5
41:11	Self-Paced Learning - 6
41:32	Image Segmentation Revisited
42:28	Outline - 6
42:39	Model Selection
42:55	Human learning revisited - 1
43:16	Human learning revisited - 2
43:18	Human learning revisited - 3
43:22	Human learning revisited - 4
43:23	Human learning revisited - 5
43:26	Human learning revisited - 6
43:31	Multiple Kernel Learning - 1
44:13	Multiple Kernel Learning - 2
44:56	Self-Paced Multiple Kernel Learning - 1
45:25	Self-Paced Multiple Kernel Learning2
45:47	SPMKL Behavior - 1
45:50	Self-Paced Multiple Kernel Learning2
45:55	SPMKL Behavior - 1
46:27	SPMKL Behavior - 2
46:53	Imputation – Iteration 1 - 2
47:12	Imputation – Iteration 3
47:16	Imputation – Iteration 1 - 2
47:21	Imputation – Iteration 3
47:31	Imputation – Iteration 6
47:37	Imputation – Iteration 10
47:37	Imputation – Iteration 10
47:42	Imputation – At Convergence
47:52	Classification Accuracy
48:36	Bounding Box Imputation
49:03	DNA Binding Motif
49:17	Conclusion I
50:03	Conclusion II
50:35	Conclusion III
50:55	The Future of Education
51:22	The Future of Machine Learning?
51:38	Acknowledgments
53:17	SPMKL Behavior - 1

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.

Lecture popularity: You need to login to cast your vote.