HIPTrack: Visual Tracking with Historical Prompts

TL;DR

HIPTrack introduces a novel Siamese-based visual tracker utilizing a historical prompt network with refined historical information, significantly improving tracking performance across multiple datasets without retraining the entire model.

Contribution

The paper presents HIPTrack, a new tracker that leverages a historical prompt network with refined historical features, enhancing performance without additional training.

Findings

Surpasses state-of-the-art on LaSOT, LaSOText, GOT-10k, and NfS datasets.

Can be integrated into existing trackers as a plug-and-play module.

Achieves significant performance gains with unchanged parameters.

Abstract

Trackers that follow Siamese paradigm utilize similarity matching between template and search region features for tracking. Many methods have been explored to enhance tracking performance by incorporating tracking history to better handle scenarios involving target appearance variations such as deformation and occlusion. However, the utilization of historical information in existing methods is insufficient and incomprehensive, which typically requires repetitive training and introduces a large amount of computation. In this paper, we show that by providing a tracker that follows Siamese paradigm with precise and updated historical information, a significant performance improvement can be achieved with completely unchanged parameters. Based on this, we propose a historical prompt network that uses refined historical foreground masks and historical visual features of the target to provide comprehensive and precise prompts for the tracker. We build a novel tracker called HIPTrack based on the historical prompt network, which achieves considerable performance improvements without the need to retrain the entire model. We conduct experiments on seven datasets and experimental results demonstrate that our method surpasses the current state-of-the-art trackers on LaSOT, LaSOText, GOT-10k and NfS. Furthermore, the historical prompt network can seamlessly integrate as a plug-and-play module into existing trackers, providing performance enhancements. The source code is available at https://github.com/WenRuiCai/HIPTrack.