Robust and Real-time Deep Tracking Via Multi-Scale Domain Adaptation

TL;DR

This paper introduces a real-time deep visual tracker that uses multi-scale domain adaptation through convolutional channel reductions, achieving high speed and state-of-the-art accuracy in benchmark tests.

Contribution

It proposes a novel feature transfer method via convolutional channel reductions for efficient and accurate deep visual tracking.

Findings

Achieves real-time tracking speeds.

Maintains high accuracy comparable to state-of-the-art methods.

Effective multi-scale adaptation improves robustness.

Abstract

Visual tracking is a fundamental problem in computer vision. Recently, some deep-learning-based tracking algorithms have been achieving record-breaking performances. However, due to the high complexity of deep learning, most deep trackers suffer from low tracking speed, and thus are impractical in many real-world applications. Some new deep trackers with smaller network structure achieve high efficiency while at the cost of significant decrease on precision. In this paper, we propose to transfer the feature for image classification to the visual tracking domain via convolutional channel reductions. The channel reduction could be simply viewed as an additional convolutional layer with the specific task. It not only extracts useful information for object tracking but also significantly increases the tracking speed. To better accommodate the useful feature of the target in different scales, the adaptation filters are designed with different sizes. The yielded visual tracker is real-time and also illustrates the state-of-the-art accuracies in the experiment involving two well-adopted benchmarks with more than 100 test videos.