The 3rd On-line learning for Computer Vision Workshop (OLCV'09) will be held in conjunction with ICCV 2009. The workshop will bring together computer vision researchers interested in providing solid foundations to this promising and challenging area.

Important Dates 

Submission of full papers	June 22, 2009
Notification of acceptance	July 20, 2009
Submission of camera ready papers	August 14, 2009
Workshop	October 3, 2009 (full day)

Scope

On-line learning is becoming essential for many real-world vision tasks. The amount of available training data is increasing rapidly, which makes off-line training times much too long. On-line algorithms process one example at a time, thus, such methods are more adequate for large data sets. In addition there are applications where on-line training is the only option (data is generated in an on-line manner). The performance of most off-line classification methods is bounded with the amount of the priori information at the beginning since they assume the learner plays no role in obtaining information about the unknown classes. However, on-line algorithms can adapt to new information. In conventional methods, the training samples are simply drawn independently from some probability distribution. On the other hand, recent on-line works show more powerful oracles. Off-line methods imply that training and testing are separate steps. Training the classifier on-line as new data arrives enables combining these stages.

Contact

If you have any comments or questions, send an email to olcv2009 AT vision.ee.ethz.ch.