Statistical Analysis of Dynamic Actions
- Creators
- Zelnik-Manor, Lihi
- Irani, Michal
Abstract
Real-world action recognition applications require the development of systems which are fast, can handle a large variety of actions without a priori knowledge of the type of actions, need a minimal number of parameters, and necessitate as short as possible learning stage. In this paper, we suggest such an approach. We regard dynamic activities as long-term temporal objects, which are characterized by spatio-temporal features at multiple temporal scales. Based on this, we design a simple statistical distance measure between video sequences which captures the similarities in their behavioral content. This measure is nonparametric and can thus handle a wide range of complex dynamic actions. Having a behavior-based distance measure between sequences, we use it for a variety of tasks, including: video indexing, temporal segmentation, and action-based video clustering. These tasks are performed without prior knowledge of the types of actions, their models, or their temporal extents.
Additional Information
© Copyright 2006 IEEE. Reprinted with permission. Manuscript received 10 Nov. 2004; revised 12 Sept. 2005; accepted 27 Jan. 2006; published online 13 July 2006. Recommended for acceptance by I.A. Essa. This work was supported by the European Commission Project IST-2000-26001 VIBES and by the Israeli Ministry of Science Grant no. 1229.Files
Name | Size | Download all |
---|---|---|
md5:01c453fa92fbb70151aa26c2ad1c3887
|
853.8 kB | Preview Download |
Additional details
- Eprint ID
- 4787
- Resolver ID
- CaltechAUTHORS:ZELieeetpami06
- Created
-
2006-09-06Created from EPrint's datestamp field
- Updated
-
2021-11-08Created from EPrint's last_modified field