Context-aware Deep Feature Compression for High-speed Visual Tracking

TL;DR

This paper introduces a context-aware deep feature compression method using multiple expert auto-encoders for high-speed visual tracking, achieving real-time performance with state-of-the-art accuracy.

Contribution

It presents a novel framework that dynamically selects the best auto-encoder based on target appearance, enabling fast and accurate tracking.

Findings

Achieves over 100 fps tracking speed.

Maintains competitive accuracy with state-of-the-art methods.

Effective feature compression with improved denoising and orthogonality loss.

Abstract

We propose a new context-aware correlation filter based tracking framework to achieve both high computational speed and state-of-the-art performance among real-time trackers. The major contribution to the high computational speed lies in the proposed deep feature compression that is achieved by a context-aware scheme utilizing multiple expert auto-encoders; a context in our framework refers to the coarse category of the tracking target according to appearance patterns. In the pre-training phase, one expert auto-encoder is trained per category. In the tracking phase, the best expert auto-encoder is selected for a given target, and only this auto-encoder is used. To achieve high tracking performance with the compressed feature map, we introduce extrinsic denoising processes and a new orthogonality loss term for pre-training and fine-tuning of the expert auto-encoders. We validate the proposed context-aware framework through a number of experiments, where our method achieves a comparable performance to state-of-the-art trackers which cannot run in real-time, while running at a significantly fast speed of over 100 fps.