機械知覚&ロボティクスグループ
中部大学

Vision Applications 国際会議

Shot Boundary Detection Using Co-occurrence of Global Motion in Video Stream

Author
Yosuke Murai, Hironobu Fujiyoshi
Publication
International Conference on Pattern Recognition, 2008

Download: PDF (Japanese)

We propose a method of shot boundary detection based on the co-occurrence of global motion in video stream. In addition to the conventional features based on appearance and local motion, we apply ST (Space- Time) patch analysis for detecting global motion in video stream. And then we perform shot boundary detection by constructing AdaBoost classifiers which represent the co-occurrence of global motion and the conventional features. Experimental results show that our method had 3.8% higher F-measure value than that of the conventional method for gradual shot boundary detection. 

前の研究 次の研究