Rethinking Convolutional Features in Correlation Filter Based Tracking

TL;DR

This paper improves visual object tracking by introducing a feature selection module that enhances feature quality, leading to better accuracy and efficiency compared to existing deep learning-based trackers.

Contribution

It proposes a novel feature selection method to improve deep feature quality, boosting tracking performance and efficiency.

Findings

Significant performance improvements over baseline trackers.

Enhanced efficiency due to reduced feature redundancy.

Competitive results against state-of-the-art methods.

Abstract

Both accuracy and efficiency are of significant importance to the task of visual object tracking. In recent years, as the surge of deep learning, Deep Convolutional NeuralNetwork (DCNN) becomes a very popular choice among the tracking community. However, due to the high computational complexity, end-to-end visual object trackers can hardly achieve an acceptable inference time and therefore can difficult to be utilized in many real-world applications. In this paper, we revisit a hierarchical deep feature-based visual tracker and found that both the performance and efficiency of the deep tracker are limited by the poor feature quality. Therefore, we propose a feature selection module to select more discriminative features for the trackers. After removing redundant features, our proposed tracker achieves significant improvements in both performance and efficiency. Finally, comparisons with state-of-the-art trackers are provided.