Particle Filter Tracking Method of Multiple Features Based Adaptive Fusion

In order to solve the poor robustness problem due to the appearance variation of the object or partial occlusion,we propose an adaptive particle filter tracking method based on the fusion of the multiple features.Two reliable features selected from the visual feature sets according to their descriptive abilities are linearly fused with the MSBRS(multiple-scale-bin-ratio-similarity) between each feature and object model.The object model is adaptively updated according to the MSBRS between the current object model and the initial model to alleviate the model drifts.Experiments show that the proposed method can robustly track the object with changes of the appearance or partly occluded.