Workshop on Visual and Contextual Learning from Annotated Images and Videos

June 25, 2009, Fontainebleau Resort, Miami Beach

VCL Workshop
Joint Workshop Links
 

 

 



Technical Program (Detailed)

9:00am Welcome Message from Chairs
9:10 am Keynote Speaker: Antonio Torralba (MIT)
Title: To think or to label ? To label: Non-parametric analysis of images and videos
9:40 am Invited Talk:  Kristen Grauman (Univ. of Texas at Austin) 
Title: Image Annotation and Continuous Visual Learning
10:10 am Coffe Break
10:30 am Keynote Speaker: David Forsyth (UIUC)
Title: Sentences and Pictures: not just more words and pictures
11:00 am Industry Talk:  Jay Yagnik (Google Research) 
Title: Computer Vision from the web with a case study of learning people annotation
11:30 am

Short Presentation
Sonal Gupta and Raymond Mooney
Using Closed Captions to Train Activity Recognizers that Improve Video Retrieval

11:45 am Short Presentation (Invited Paper)
Timothee Cour, Benjamin Sapp , Akash Nagle, Ben Taskar 
Talking Pictures: Temporal Grouping and Dialog‐Supervised Person Recognition in Video
12:00 noon Spotlights (5 min each) - Joint with ViSU'09

Oksana Yakhnenko, Vasant Honavar 
Multiple Label Prediction for Image Annotation with Multiple Kernel Correlation Models

Ozge Oztimur Karadağ, Fatoş T. Yarman Vural 
HANOLISTIC: A Hierarchical Automatic Image Annotation System Using Holistic Approach 
12:30-2:10 Poster Session* - Joint with ViSU'09

Oksana Yakhnenko, Vasant Honavar 
Multiple Label Prediction for Image Annotation with Multiple Kernel Correlation Models
 
Ozge Oztimur Karadağ, Fatoş T. Yarman Vural 
HANOLISTIC: A Hierarchical Automatic Image Annotation System Using Holistic Approach 

Sonal Gupta and Raymond Mooney
Using Closed Captions to Train Activity Recognizers that Improve Video Retrieval

Hong Lu, Yingbin Zheng, Xiangyang Xue, Yuejie Zhang 
Content and Context‐Based Multi‐Label Image Annotation

Anup Doshi, Mohan M. Trivedi
Head and Gaze Dynamics in Visual Attention and Context Learning

Hitoshi Habe, Hidehito Nakagawa, Masatsugu Kidode 
Efficient Acquisition of Human Existence Priors from Motion Trajectories
2:10 pm ViSU'09
5:30 pm Concluding Remarks

*Poster Board Sizes: Those presenting posters will hang their posters on tackboards. The usable space on these tack boards is approximately 43.5 inches high x 91.25 inches wide.

Invited Speakers:

David Forsyth (UIUC)
   

Antonio Torralba (MIT)
   

Kristen Grauman (UT Austin)
   

Jay Yagnik (Google Research)

 

 

Webmaster: Abhinav Gupta