Multi-scale correlation tracking with convolutional features

Abstract

Feature extractor plays an important role in visual tracking due to the changing appearance of the object. In this paper, we propose a novel approach in correlation filter framework, which decomposes the task of tracking into translation and scale estimation. We employ two correlation filters with hierarchical convolutional features to estimate the translation. Furthermore, we use a discriminative correlation filter with histogram of oriented gradient features to handle scale variations. Extensive experiments are performed on a large-scale benchmark challenging dataset. And the results show that the proposed algorithm outperforms state-of-the-art tracking methods in accuracy and robustness.