Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

UETrack: A Unified and Efficient Framework for Single Object Tracking

Ben Kang, Jie Zhao, Xin Chen, Wanting Geng, Bin Zhang, Lu Zhang, Dong Wang, Huchuan Lu

TL;DR

UETrack introduces two key components: a Token-Pooling-based Mixture-of-Experts mechanism that enhances modeling capacity through feature aggregation and expert specialization, and a Target-aware Adaptive Distillation strategy that selectively performs distillation based on sample characteristics, reducing redundant supervision and improving performance.

Abstract

With growing real-world demands, efficient tracking has received increasing attention. However, most existing methods are limited to RGB inputs and struggle in multi-modal scenarios. Moreover, current multi-modal tracking approaches typically use complex designs, making them too heavy and slow for resource-constrained deployment. To tackle these limitations, we propose UETrack, an efficient framework for single object tracking. UETrack demonstrates high practicality and versatility, efficiently handling multiple modalities including RGB, Depth, Thermal, Event, and Language, and addresses the gap in efficient multi-modal tracking. It introduces two key components: a Token-Pooling-based Mixture-of-Experts mechanism that enhances modeling capacity through feature aggregation and expert specialization, and a Target-aware Adaptive Distillation strategy that selectively performs distillation based on sample characteristics, reducing redundant supervision and improving performance. Extensive experiments on 12 benchmarks across 3 hardware platforms show that UETrack achieves a superior speed-accuracy trade-off compared to previous methods. For instance, UETrack-B achieves 69.2% AUC on LaSOT and runs at 163/56/60 FPS on GPU/CPU/AGX, demonstrating strong practicality and versatility. Code is available at https://github.com/kangben258/UETrack.

UETrack: A Unified and Efficient Framework for Single Object Tracking

TL;DR

UETrack introduces two key components: a Token-Pooling-based Mixture-of-Experts mechanism that enhances modeling capacity through feature aggregation and expert specialization, and a Target-aware Adaptive Distillation strategy that selectively performs distillation based on sample characteristics, reducing redundant supervision and improving performance.

Abstract

With growing real-world demands, efficient tracking has received increasing attention. However, most existing methods are limited to RGB inputs and struggle in multi-modal scenarios. Moreover, current multi-modal tracking approaches typically use complex designs, making them too heavy and slow for resource-constrained deployment. To tackle these limitations, we propose UETrack, an efficient framework for single object tracking. UETrack demonstrates high practicality and versatility, efficiently handling multiple modalities including RGB, Depth, Thermal, Event, and Language, and addresses the gap in efficient multi-modal tracking. It introduces two key components: a Token-Pooling-based Mixture-of-Experts mechanism that enhances modeling capacity through feature aggregation and expert specialization, and a Target-aware Adaptive Distillation strategy that selectively performs distillation based on sample characteristics, reducing redundant supervision and improving performance. Extensive experiments on 12 benchmarks across 3 hardware platforms show that UETrack achieves a superior speed-accuracy trade-off compared to previous methods. For instance, UETrack-B achieves 69.2% AUC on LaSOT and runs at 163/56/60 FPS on GPU/CPU/AGX, demonstrating strong practicality and versatility. Code is available at https://github.com/kangben258/UETrack.
Paper Structure (12 sections, 4 equations, 7 figures, 7 tables)

This paper contains 12 sections, 4 equations, 7 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. UETrack
  4. Overall Architecture
  5. Token-Pooling-based MoE
  6. Target-aware Adaptive Distillation
  7. Training Objective
  8. Experiments
  9. Implementation Details
  10. State-of-the-Art Comparisons
  11. Ablation and Analysis
  12. Conclusion

Figures (7)

  • Figure 1: UETrack vs. Other Trackers. (a) compares UETrack with current efficient and multi-modal trackers; (b) presents a comparison of speed-accuracy trade-offs on the Jetson AGX.
  • Figure 2: Architecture of UETrack. The training pipeline consists of a teacher model, a student model, and an Adaptive Net for adaptive distillation. During inference, only the student model is used, with TP-MoE as the core component to enhance multi-modal modeling.
  • Figure 3: TP-MoE architecture diagram.
  • Figure 4: Architecture of Adaptive Net.
  • Figure 5: EAO rank plots on VOT2021 Real-time.
  • ...and 2 more figures