HighlightNet: Highlighting Low-Light Potential Features for Real-Time UAV Tracking

TL;DR

HighlightNet is a Transformer-based enhancer that improves low-light UAV tracking by highlighting potential features, adapting to illumination changes, and focusing on relevant regions, thereby enhancing tracking accuracy and human perception in challenging conditions.

Contribution

This work introduces HighlightNet, a novel adaptive low-light enhancer utilizing Transformer and pixel-level masks, specifically designed for real-time UAV tracking in low-light environments.

Findings

HighlightNet outperforms existing low-light enhancers on benchmarks.

It improves UAV tracking accuracy in nighttime conditions.

Real-world tests confirm its practicality and efficiency.

Abstract

Low-light environments have posed a formidable challenge for robust unmanned aerial vehicle (UAV) tracking even with state-of-the-art (SOTA) trackers since the potential image features are hard to extract under adverse light conditions. Besides, due to the low visibility, accurate online selection of the object also becomes extremely difficult for human monitors to initialize UAV tracking in ground control stations. To solve these problems, this work proposes a novel enhancer, i.e., HighlightNet, to light up potential objects for both human operators and UAV trackers. By employing Transformer, HighlightNet can adjust enhancement parameters according to global features and is thus adaptive for the illumination variation. Pixel-level range mask is introduced to make HighlightNet more focused on the enhancement of the tracking object and regions without light sources. Furthermore, a soft truncation mechanism is built to prevent background noise from being mistaken for crucial features. Evaluations on image enhancement benchmarks demonstrate HighlightNet has advantages in facilitating human perception. Experiments on the public UAVDark135 benchmark show that HightlightNet is more suitable for UAV tracking tasks than other SOTA low-light enhancers. In addition, real-world tests on a typical UAV platform verify HightlightNet's practicability and efficiency in nighttime aerial tracking-related applications. The code and demo videos are available at https://github.com/vision4robotics/HighlightNet.