User-Centric Retrieval of Visual Surveillance Content

2nd International Conference on Semantic and Digital Media Technologies

Jerome Meessen ; Xavier Desurmont ; Christophe De Vleeschouwer ; Jean-Franc¸ois Delaigle

Others  :  http://CEUR-WS.org/Vol-300/p09.pdf
PID  :  21746

来源: CEUR

【 摘 要 】

An interactive retrieval method adapted to surveillance video is presented. The approach is formulated as an iterative SVM classification and builds upon the two major specificities of the surveillance context, namely the multiple instance nature of the data and the reduced number of training examples the user can provide at each round. The later issue is solved thanks to a new adaptive active learning strategy as well as an intuitive graphical user interface. The system has been validated on both synthetic and real datasets.