Target Tracking Method Based on Adaptive Structured Sparse Representation With Attention

Abstract

Considering the problems of motion blur, partial occlusion and fast motion in target tracking, a target tracking method based on adaptive structured sparse representation with attention is proposed. Under the framework of particle filtering, the performance of high-quality templates is enhanced through an attention mechanism. Structure sparseness is used to build candidate target sets and sparse models between candidate samples and local patches of target templates. Combined with the sparse residual method, reconstruction error is reduced. After optimally solving the model, the particle with the highest similarity is selected as the prediction target. The most appropriate scale is selected according to the multiscale factor method. Experiments show that the proposed algorithm has a strong performance when dealing with motion blur, fast motion, partial occlusion.

Particle Filter Based Video Object Tracking Scheme With Target Remodeling and Reinitialization and its Hardware Implementation Using Raspberry Pi

“…Based on different issues of the scene complexity, different methods of tracking have been proposed in literature such as feature-based tracking [7][8][9][10], model-based tracking [11][12][13][14][15][16][17], region-based tracking [18], and deformable template-based tracking [19][20][21]. It has been observed that particle filter in feature space provides a clear understanding and effective analysis of the target object in a scene.…”