VideoMap: Interactive Search System by MCG-ICT-CAS*
Juan Cao, Yongdong Zhang, Bailan Feng, Lei Bao, Ling Pang, Jintao Li, Ke Gao, Xiao Ying Wu, Hon-Ttao Xie, Wei Zhang, Zhendong Mao · 2009
This paper describes the highlights of our interactive search system VideoMap for TRECVID 2009. To enhance the efficiency, the system has a map based displaying interface, which gives the user a global view about the similarity relationships among the whole video collection, and provides an active annotating manner to quickly localize the potential positive samples. Meanwhile, the system has powerful multiple modality feedback strategies, including the visual-based feedback, concept-based feedback and community-based feedback. These feedback algorithms can flexibly transform between each other by the automatic optimizing strategy. Finally the multi-strategies feedback achieves the best performance with MAP of 0.186.