Efficient and cost-effective techniques for browsing and indexing large video databases
SIGMOD '00 Proceedings of the 2000 ACM SIGMOD international conference on Management of data
Content-based video similarity model
MULTIMEDIA '00 Proceedings of the eighth ACM international conference on Multimedia
Wavelet-based video indexing and querying
Multimedia Systems - Special section on video libraries
Taxonomy of directing semantics for film shot classification
IEEE Transactions on Circuits and Systems for Video Technology
Hi-index | 0.00 |
Since an entire video stream is too coarse as a level of abstraction, it is decomposed into a number of shots in general. A shot, which is defined as a collection of frames recorded from a single camera operation, is a basic unit for indexing video data. The existing video indexing models try to index these shots using either low-level (i.e., color, motion), or high-level (i.e., human interpretations) features. Recently, hybrid approach to combine these low-level and high-level features for indexing purpose has been proposed to take advantages from both of them. In this paper, we extend our previous works for shot boundary detecting and indexing, and propose a new shot indexing technique to be fit into the hybrid environment. It supports in-depth classification of various kinds of shots according to camera motions and object movements. The results from preliminary experiments can confirm the effectiveness of the proposed techniques.