Abstrait

Non-rigid object tracking via discriminative features

Qian Wang, Qingxuan Shi

Non-rigid objects are typically complex and difficult to track due to the appearance change caused by geometric changes. In this paper, we model the appearance of non-rigid objects by discriminative features which are adaptively selected according to their descriptive ability. To adapt to the geometric changes, we use a deformable rectangle to represent the object, and use Markov Chain Monte Carlo-based Particle Filter (MCMCPF) to estimate the state of the object in a restricted four-dimensional space. Experimental results show that the proposed tracking algorithm has ideal performance.