2011 CVPR Workshop on
Large Scale Learning for Vision
Monday June 20, 2011
Welcome to the first large scale learning in vision workshop!

Jump to the schedule/slides.

Successful meeting thanks to all 250+ attendees!
(discussion in the morning section)


A recent theme throughout science is the advent of "big data" -- leveraging huge datasets to tackle fundamental problems. Computer Vision is also experiencing this paradigm shift, with large annotated image and video datasets becoming available to researchers. These have the potential to revolutionize everything from low level vision to object detection to scene understanding. But to realize this potential, researchers must address the fundamental computational problems inherent with such large datasets. Concurrently, Machine Learning researchers are exploring this "big data" paradigm, developing theory and techniques that can be applied to a range of data modalities. The goal of this workshop is to bring together leaders in the theory and practice of large scale machine learning with computer vision researchers who have begun to explore the challenge and promise of large amounts of data.
  • Identify large scale problems in computer vision
  • Present large scale machine learning and optimization techniques, distinct from those used in computer vision
  • Discuss new directions provided by the problems, data, and techniques from both sides
  • Encourage vision researchers to expand their repertoire of approaches
  • Remind machine learning researchers that vision is an exciting problem and to encourage work that benefits our community!
The format of the workshop will consist of invited speakers and panel discussions. There are no paper submissions.
8:00-9:00Breakfast Buffet 
David Forsyth
University of Illinois at Urbana-Champain
More Words and Bigger Pictures: Where could large-scale learning take us?
Eric Xing
Carnegie Mellon University
Sparsity and Learning Large Scale Models
Jason Weston
Google Incorporated
Large Scale Image Annotation: Learning to Rank with Joint Word-Image Embeddings
11:20-11:40Fei-Fei Li
Stanford University
Building the Forest: Large Scale Data and Modeling in Computer Vision
11:40-12:00Discussion ( All ) 
12:00-1:20Lunch Buffet 
1:20-1:40James Hays
Brown University
Scene understanding and representation at Internet-scale.
1:40-2:00Kristen Grauman
University of Texas at Austin
Live Large-Scale Active Learning
2:00-2:20Svetlana Lazebnik
University of North Carolina, Chapel Hill
Large-scale nonparametric image parsing
2:20-2:50Discussion ( All ) 
2:50-3:00PrizesMost Controversial, Most Questions, etc.