Bitstream-based overlap analysis for multi-view distributed video coding

In this paper, we develop a low complexity algorithm for spatial overlap detection and characterization that operates directly on the bitstream of motion-JPEG compressed video. Its low complexity and the fact that it does not require video decoding at the sensor nodes makes it well suited to multi-view distributed video coding applications for wireless sensor networks. In addition to determining overlap, we also show using the MERL ballroom database that the proposed algorithm can accurately identify and localize non-common foreground objects (e.g., dancers) within the common overlap region of two camera views.