An Improved Compressive Tracking Algorithm Based on LBP

This paper presents an improved compressive tracking algorithm based on LBP (Local Binary Pattern). Compressive Tracking is a simple and effective tracking algorithm, however there are a number of problems that need to be addressed. In our algorithm, the target is represented with Haar-like features generated from three channels of RGB LBP image with completely random location and channel rather than from a single channel. We also propose a new search strategy to update the search window instead of a fixed tracking window. The improved algorithm is tested on various video sequences and the results show that the algorithm achieves stable tracking on moved target and has a good performance compared with the state-of-the-art algorithm in terms of efficiency and accuracy.