Abstract—long term occlusion is a most important challenge in any multiple objects tracking system. This paper presents a literature survey of an object tracking algorithms in a fixed camera situation that have been used by others to address the long term occlusion problem. Based on this assessment of the state of the art, this paper identify what appears to be the most promising algorithm for long term occlusion that was succeeds in handling interacting objects of similar appearance without any strong assumptions on the characteristics of the tracked objects. But this algorithm failed to handle objects of too complex shapes and appearance, and the tracking results was affected by successfully of background subtraction. This paper presents a proposed solution to these failed points. |
