|
| | | |
Role of Edge Detection in Video Semantics
Lee, M.D., Nepal, S. and Srinivasan, U.
The semantic gap or semantic chasm is a well-known problem in content-based image and video retrieval. To address this problem, many techniques have been proposed in the literature. A more common approach is the use of low-level features such as colour, texture and shape for semantic analysis. Our focus in this paper is on the edge feature, which has not been exploited to the same extent as other low-level features for semantic analysis. In this paper, we present an algorithm for edge detection, and illustrate the usage of edges for semantic analysis of video content. We first propose an algorithm for detecting edges within video frames directly on the MPEG format without a decompression process. The algorithm is based on a spatial-domain synthetic edge model, which is defined using interrelationship of two DCT edge features: horizontal and vertical. We use a multi-step approach to classify video sequences into meaningful semantic segments such as 'goal', 'foul', and 'crowd' in basketball games using the 'edgeness' criteria. We then show how an audio feature ('whistle') can be used as a filter to enhance edgebased semantic classification fro sports videos. |
Cite as: Lee, M.D., Nepal, S. and Srinivasan, U. (2003). Role of Edge Detection in Video Semantics. In Proc. Pan-Sydney Area Workshop on Visual Information Processing (VIP2002), Sydney, Australia. CRPIT, 22. Jin, J. S., Eades, P., Feng, D. D. and Yan, H., Eds. ACS. 59. |
(from crpit.com)
(local if available)
|
|