Discriminative Genre-Independent Audio-Visual Scene Change Detection
| Citation: | Wilson, K.W.; Divakaran, A., "Discriminative Genre-Independent Audio-Visual Scene Change Detection", SPIE Conference on Multimedia Content Access: Algorithms and Systems III, Vol. 7255, , January 2009 (SPIE Digital Library) |
| MERL Report: | TR2009-001 |
| MERL Contact: | Kevin W. Wilson |
SVM Classifier Framework
We present a technique for genre-independent scene-change detection using audio and video features in a discriminative support vector machines (SVM) framework. This work builds on our previous work by adding a video feature based on the MPEG-7 "scalable color" descriptor. Adding this feature imporoves our detection rate over all genres by 5% to 15% for a fixed false positive rate of 10%. We also find that the genres that benefit the most are those with which the previous audio-only was least effective.
 | Read the full technical report (PDF: 246.4 kB) |