Particle filter algorithm based on information-shared mechanism and its application to visual tracking
基于信息分享机制的粒子滤波算法及其在视觉跟踪中的应用

To deal with the weight-degeneracy and computation efficiency in particle filter, we propose a novel particle filter algorithm based on the information-shared mechanism. This method combines particle swarm optimization and ant colony optimization to update particles to fully share the population information. The particles diversity is recovered and the estimation precision is improved. The convergence of this algorithm is analyzed. Visual tracking experiments show that the proposed algorithm realizes both single-object tracking and multi-object tracking with fewer particles and better comprehensive tracking-performance than classic particle filters and the particle filter based on particle-swarm optimization.