cs.CVNov 3, 2026

TAQ-MOTR: Tracking-Aware Detection Query Formation for End-to-End Multiple Object Tracking

Authors: Hanyang Liu, Hong Zhang, Ding Yuan, Yang Han, Yifan Yang

Organizations: School of Astronautics, Beihang University, 37 Xueyuan Road, Haidian District, Beijing 100191, China · Key Laboratory of Spacecraft Design Optimization and Dynamic Simulation Technology, Ministry of Education · State Key Laboratory of High-Efficiency Reusable Aerospace Transportation Technology · School of Software, Beihang University, 37 Xueyuan Road, Haidian District, Beijing 100191, China

Abstract

Tracking-by-propagation (TBP) methods perform end-to-end multi-object tracking by propagating track queries across frames, jointly handling detection and identity association without heuristic post-processing while naturally modeling temporal information. However, as detection and association are tightly coupled, detection queries in TBP are expected to discover currently untracked objects and provide complementary current-frame evidence for challenging tracks whose queries and identities are retained. Existing methods largely inherit detection-query formation from single-frame detection and construct such queries before the decoder with little awareness of trajectory states. Consequently, currently untracked objects may be missed, while challenging tracked objects with reduced visibility under occlusion may lack complementary evidence, causing association failures or duplicate responses. To address these issues, we propose TAQ-MOTR, a tracking-aware detection query formation framework that employs a pre-decoder Hybrid Query Generator (HQG) to fuse current-frame image features, track memory, and learnable memory. By conditioning detection queries on trajectory states, TAQ-MOTR adapts them to \rev{the discovery of currently untracked objects and tracked-object complementation We further introduce a training-only auxiliary detection branch whose one-to-one label assignment is decoupled from that of the main tracking branch, providing additional detection supervision to strengthen the shared HQG and decoder for the discovery of currently untracked objects. TAQ-MOTR improves performance on DanceTrack, SportsMOT, and BFT, achieving 72.0 HOTA, 63.1 AssA, and 76.6 IDF1 on the DanceTrack test set. Code is available at https://github.com/Magicmiya/TAQ-MOTR

Figures & tables

Explore similar work

CardsList
  1. Out of Sight, Out of Track: Adversarial Attacks on Propagation-based Multi-Object Trackers via Query State Manipulation

    Apr 1, 2026Halima Bouzidi, Haoyu Liu, Yonatan Gizachew Achamyeleh +2Adversarial AttacksMulti-Object Tracking

  2. Beyond the Survey: A Systematic Empirical Study of Detection and Association in Visual MOT

    Sep 14, 2026Linh Van Ma, Juhua Hu, Wei Cheng +2Multi-Object TrackingData Association