Occlusion detection in moving cameras
Tarkan Aydın, A. Tanju Erdem · 2012
In this work, we propose a novel method for detection of moving or static objects which are not part of the background in an image sequence captured by a moving camera. The method is composed of two stages, namely, construction of a background model and detection of occluding objects. The constructed background model makes it possible to detect both static and moving occluding objects. Experiments on the real scenes show that the proposed method works effectively and produces visually correct results.