Columbia University TRECVID-2006 Video Search and High-Level Feature Extraction

 A_CL1_1: choose the best-performing classifier for each concept from all the following runs and an event detection method.  A_CL2_2: (visual-based) choose the best-performing visual-based classifier for each concept from runs A_CL4_4, A_CL5_5, and A_CL6_6.  A_CL3_3: (visual-text) weighted average fusion of visual-based classifier A_CL4_4 with a text classifier.  A_CL4_4: (visual-based) average fusion of A_CL5_5 with a lexicon-spatial pyramid matching approach that incorporates local feature and pyramid matching.  A_CL5_5:(visual-based) average fusion of A_CL6_6 with a context-based concept fusion approach that incorporates inter-conceptual relationships to help detect individual concepts.  A_CL6_6: (visual-based) average fusion of two SVM baseline classification results for each concept: (1) a fused classification result by averaging 3 single SVM classification results. Each SVM classifier uses one type of color/texture/edge features and is trained on the whole training set with 40% negative samples; (2) a single SVM classifier using color and texture trained on the whole training set (90 videos) with 20% negative samples.