Targets Tracking in the Crowd

Conventional video surveillance systems often have several shortcomings. First, target detection can’t be accurate under the light variation environment or clustering backgrounds. Second, multiple targets tracking become difficult on a crowd scene because the split/merge and occlusions among the tracked targets occur frequently and irregularly. Third, it is difficult to the partition the tracked targets from a merged image blob and then the target tracking may be inaccurate. In this chapter, the methods for targets detection and tracking in the crowd are addressed. In general, the methods for targets tracking in the crowd can be categorized into the blob-based and point-based methods. The blob-based methods detect and track the targets based on the appearance models; while the point-based methods detect and track the moving targets with the reliable feature points.