event thumbnail image
Sessions

Deep Learning from Temporal Coherence in Video

author: Ronan Collobert, NEC Laboratories America, Inc.

Description

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.

You might be experiencing some problems with Your Video player.
Slides
0:00 Deep Learning from Temporal Coherence in Video
0:15 The Goal
1:32 Embedding Algorithm
3:22 Embedding Algorithm: Applications
5:06 Video: Temporal Coherence
6:05 Leveraging Temporal Coherence
6:42 Algorithm
7:23 Previous Work: Semi-supervised Learning
8:39 Bad Metric: Euclidean Distance - 1
9:02 Bad Metric: Euclidean Distance - 2
9:09 Bad Metric: Euclidean Distance - 3
9:12 Bad Metric: Euclidean Distance - 4
9:55 Cluster Assumption
10:36 Previous Work: Semi-supervised Learning
11:03 Previous Work: Temporal Coherence
12:21 Experiments: COIL 100 Setup
12:54 Experiments: Coil 100
13:20 Experiments: Coil 100-Like
13:51 Experiments: Animal Set
14:03 Experiments: COIL 100 Performance
15:41 Experiments: AT&T’s ORL Face Dataset
16:01 Experiments: Simple ORL Experiment
16:47 Conclusion

Lecture rating

People found this lecture:
Worth seeing
because it is:
 Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

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.

Link this page

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

Write your own review or comment:

make sure you have javascript enabled or clear this field: