作者: Toon Calders , Hoang Thanh Lam , Elisa Fromont , TU Eindhoven , Baptiste Jeudy
DOI:
关键词:
摘要: We investigate how mining top-k largest tiles in a data stream under the sliding window model can be useful for (real-time) analysis of videos and, particular, tracking. first explain track- ing problem cast into pattern problem. then show some preliminary results on tracking particular context where both objects and camera are moving user does not specify regions interest frames videos.