UTPTrack: Towards Simple and Unified Token Pruning for Visual Tracking

• URL: http://arxiv.org/abs/2602.23734v1
• Date: Fri, 27 Feb 2026 06:58:09 GMT
• Title: UTPTrack: Towards Simple and Unified Token Pruning for Visual Tracking
• Authors: Hao Wu, Xudong Wang, Jialiang Zhang, Junlong Tong, Xinghao Chen, Junyan Lin, Yunpu Ma, Xiaoyu Shen,
• Abstract summary: One-stream Transformer-based trackers achieve advanced performance in visual object tracking but suffer from significant computational overhead.<n>We introduce UTPTrack, a simple and Unified Token Pruning framework that, for the first time, jointly compresses all three components.<n>Extensive evaluations on 10 benchmarks demonstrate that UTPTrack achieves a new state-of-the-art in the accuracy-efficiency trade-off for pruning-based trackers.
• Score: 23.83535022949326
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: One-stream Transformer-based trackers achieve advanced performance in visual object tracking but suffer from significant computational overhead that hinders real-time deployment. While token pruning offers a path to efficiency, existing methods are fragmented. They typically prune the search region, dynamic template, and static template in isolation, overlooking critical inter-component dependencies, which yields suboptimal pruning and degraded accuracy. To address this, we introduce UTPTrack, a simple and Unified Token Pruning framework that, for the first time, jointly compresses all three components. UTPTrack employs an attention-guided, token type-aware strategy to holistically model redundancy, a design that seamlessly supports unified tracking across multimodal and language-guided tasks within a single model. Extensive evaluations on 10 benchmarks demonstrate that UTPTrack achieves a new state-of-the-art in the accuracy-efficiency trade-off for pruning-based trackers, pruning 65.4% of vision tokens in RGB-based tracking and 67.5% in unified tracking while preserving 99.7% and 100.5% of baseline performance, respectively. This strong performance across both RGB and multimodal scenarios underlines its potential as a robust foundation for future research in efficient visual tracking. Code will be released at https://github.com/EIT-NLP/UTPTrack.

Related papers

• UETrack: A Unified and Efficient Framework for Single Object Tracking [46.50641228786134]
  UETrack is an efficient framework for single object tracking.<n>It efficiently handles multiple modalities including RGB, Depth, Thermal, Event, and Language.<n>It achieves a superior speed-accuracy trade-off compared to previous methods.
  arXiv  Detail & Related papers  (2026-03-02T03:32:30Z)
• UniTrack: Differentiable Graph Representation Learning for Multi-Object Tracking [5.241700353040585]
  UniTrack is a plug-and-play graph-theoretic loss function designed to enhance multi-object tracking (MOT)<n>We show up to 53% reduction in identity switches and 12% IDF1 improvements across challenging benchmarks, with peak performance gains 9.7% MOTA on SportsMOT.
  arXiv  Detail & Related papers  (2026-02-04T20:44:16Z)
• Two-stream Beats One-stream: Asymmetric Siamese Network for Efficient Visual Tracking [54.124445709376154]
  We propose a novel asymmetric Siamese tracker named textbfAsymTrack for efficient tracking.<n>Building on this architecture, we devise an efficient template modulation mechanism to inject crucial cues into the search features.<n>Experiments demonstrate that AsymTrack offers superior speed-precision trade-offs across different platforms.
  arXiv  Detail & Related papers  (2025-03-01T14:44:54Z)
• Exploring Dynamic Transformer for Efficient Object Tracking [58.120191254379854]
  We propose DyTrack, a dynamic transformer framework for efficient tracking.<n>DyTrack automatically learns to configure proper reasoning routes for various inputs, gaining better utilization of the available computational budget.<n>Experiments on multiple benchmarks demonstrate that DyTrack achieves promising speed-precision trade-offs with only a single model.
  arXiv  Detail & Related papers  (2024-03-26T12:31:58Z)
• Single-Model and Any-Modality for Video Object Tracking [85.83753760853142]
  We introduce Un-Track, a Unified Tracker of a single set of parameters for any modality. To handle any modality, our method learns their common latent space through low-rank factorization and reconstruction techniques. Our Un-Track achieves +8.1 absolute F-score gain, on the DepthTrack dataset, by introducing only +2.14 (over 21.50) GFLOPs with +6.6M (over 93M) parameters.
  arXiv  Detail & Related papers  (2023-11-27T14:17:41Z)
• Revisiting Color-Event based Tracking: A Unified Network, Dataset, and Metric [53.88188265943762]
  We propose a single-stage backbone network for Color-Event Unified Tracking (CEUTrack), which achieves the above functions simultaneously. Our proposed CEUTrack is simple, effective, and efficient, which achieves over 75 FPS and new SOTA performance.
  arXiv  Detail & Related papers  (2022-11-20T16:01:31Z)
• Joint Feature Learning and Relation Modeling for Tracking: A One-Stream Framework [76.70603443624012]
  We propose a novel one-stream tracking (OSTrack) framework that unifies feature learning and relation modeling. In this way, discriminative target-oriented features can be dynamically extracted by mutual guidance. OSTrack achieves state-of-the-art performance on multiple benchmarks, in particular, it shows impressive results on the one-shot tracking benchmark GOT-10k.
  arXiv  Detail & Related papers  (2022-03-22T18:37:11Z)
• Learning Dynamic Compact Memory Embedding for Deformable Visual Object Tracking [82.34356879078955]
  We propose a compact memory embedding to enhance the discrimination of the segmentation-based deformable visual tracking method. Our method outperforms the excellent segmentation-based trackers, i.e., D3S and SiamMask on DAVIS 2017 benchmark.
  arXiv  Detail & Related papers  (2021-11-23T03:07:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.