Motion Clustering For Similar Video Segments Mining

Kexue Dai,Guohui Li,Defeng Wu
DOI: https://doi.org/10.1109/MMMC.2006.1651368
2006-01-01
Abstract:To discover similar video segments from surveillance video sequence, a new approach is proposed for clustering motion dato of moving objects. A sample background subtraction algorithm is used to get the binary mask of moving objects for segmenting the video sequence captured by fixed camera. Then a mixture of hidden Markov models (HMMs) using the expectation-maximization (EM) scheme is fitted to the motion data extracted from the binary mask. Unlike previous literatures using k-means where every observed data set only assigned to a single HMM, the proposed approach allows every video segment to belong to more than a single HMM with some probability. Expermients with real data demonstrate the benefit when there is more "overlap" the processes generating the data. The promising potential of HMM-based motion clustering for mining similar video segments from surveillance video is also indicated by the experimental results.
What problem does this paper attempt to address?