CSDL Home D DANTE 1999 Database Applications in Non-Traditional Environments, International Symposium on
Nov. 28, 1999 to Nov. 30, 1999
Pu-Jien Cheng , National Chiao Tung University
Wei-Pang Yang , National Chiao Tung University
This paper presents a new conceptual model for representing visual information about moving objects in video data. Based on available automatic scene segmentation and object tracking algorithms, the proposed model calculates object motions at various levels of semantic granularity. It represents trajectory, color and dimensions of a single moving object and the directional and topological relations among multiple objects over a time interval. To facilitate query processing, there are two optimal approximate matching algorithms designed to match time-series visual features of moving objects. Experimental results indicate that the proposed algorithms outperform the conventional subsequence-matching methods substantially in the similarity between the two trajectories.
content-based retrieval, video data modeling, approximate string matching and spatio-temporal representation
Pu-Jien Cheng, Wei-Pang Yang, "Querying Video Contents by Motion Example", DANTE, 1999, Database Applications in Non-Traditional Environments, International Symposium on, Database Applications in Non-Traditional Environments, International Symposium on 1999, pp. 287, doi:10.1109/DANTE.1999.844972