Target motion compensation with optical flow clustering during visual tracking
Cheng‐Ming Huang, Ming-Hong Hung · 2014
In this paper, we proposed a tracking system, which is based on the particle filtering with the optical flow clustering, to overcome the tracking problem of target motion with the violent camera displacement or the fast target movement. The multi-thread processing algorithm is proposed to reliably obtain the optical flow information uniformly diffused in the whole image frame. By clustering and analyzing the optical flow on images, the target motion can be estimated to compensate the regular target motion model and then efficiently draw the particles for tracking target. Experimental results show that our proposed tracking methods can efficiently overcome the target tracking under various camera moving situations with a small number of particles.