Deep Learning from Temporal Coherence in Video

author: Ronan Collobert, NEC Laboratories America, Inc.
published: Aug. 26, 2009,   recorded: June 2009,   views: 6699


Related Open Educational Resources

Related content

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.


This work proposes a learning method for deep architectures that takes advantage of sequential data, in particular from the temporal coherence that naturally exists in unlabeled video recordings. That is, two successive frames are likely to contain the same object or objects. This coherence is used as a supervisory signal over the unlabeled data, and is used to improve the performance on a supervised task of interest. We demonstrate the effectiveness of this method on some pose invariant object and face recognition tasks.

See Also:

Download slides icon Download slides: icml09_collobert_dlt_01.pdf (2.7┬áMB)

Help icon Streaming Video Help

Link this page

Would you like to put a link to this lecture on your homepage?
Go ahead! Copy the HTML snippet !

Reviews and comments:

Comment1 coppermine, March 15, 2010 at 10:10 a.m.:

This is nice presentation because it chooses bit different way of learning from large unlabeled datasets, especially when they are sequential as videos. The results are attractive... however, the presentation seemed little bit too high level. It would be nice if more technical details would be involved.

Write your own review or comment:

make sure you have javascript enabled or clear this field: