Genetic CONDENSATION for motion tracking

Tracking is a particularly important issue in human motion analysis since it serves as a means to prepare data for pose estimation and action recognition. The CONDENSATION algorithm is a kind of conditional density propagation method for motion tracking. This algorithm combines factored sampling with learned dynamic models to propagate an entire probability distributes for object position and shape over time. It can accomplish highly robust tracking of object motion. However, it usually requires a large number of samples to ensure a fair maximum likelihood estimation of the current state. The important problem of the CONDENSATION algorithm is to choose proper samples to approach the actual samples position. In this paper, we use the mutation and crossover operators of the genetic algorithm to find more appropriate samples by calculating weights. Accordingly, we can solve the heavy demand of samples in the CONDENSATION algorithm. Eventually, we can improve robustness, accuracy and flexibility in CONDENSATION for visual tracking.