mlsanityData to 14 Oct 2026

Referring Multi-Object Tracking in Moving-Camera Videos via Global Motion Compensation

arXiv 2610.04185 ↗First posted 3 Oct 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Referring multi-object tracking (RMOT) takes a video and a language expression as input and tracks all referred objects. Many tracking requirements involve how an object moves rather than how it appears. However, in a video captured by a moving camera, a parked vehicle may appear to move, while a moving vehicle may show little displacement. Existing RMOT methods relate motion with text but do not explicitly remove camera-induced motion. In this paper, we propose extracting residual motion across frames by estimating camera motion in driver-view videos and compare motion characteristics with the query expression. We consider the motion-matching extent and integrate it with the RMOT method's prediction result through late fusion. In the evaluation, we verify the performance gain of taking the motion compensation module as a plug-in across different RMOT hosts.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 13 of its 19 references that are in the map. First posted on arXiv 3 Oct 2026.

Builds on

13 of 19 references.

References · 13

  1. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  2. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  3. 2020HOTA: A Higher Order Metric for Evaluating Multi-Object Trackingcited by 552
  4. 2022BoT-SORT: Robust Associations Multi-Pedestrian Trackingcited by 320
  5. 2023Referring Multi-Object Trackingcited by 93
  6. 2023UCMCTrack: Multi-Object Tracking with Uniform Camera Motion Compensationcited by 44
  7. 2023iKUN: Speak to Trackers without Retrainingcited by 32
  8. 2024Bootstrapping Referring Multi-Object Trackingcited by 22
  9. 2025ReferGPT: Towards Zero-Shot Referring Multi-Object Trackingcited by 11
  10. 2025MEX: Memory-efficient Approach to Referring Multi-Object Trackingcited by 4
  11. 2024Ego-Motion Aware Target Prediction Module for Robust Multi-Object Trackingcited by 1
  12. 2025Rethinking Two-Stage Referring-by-Tracking in Referring Multi-Object Tracking: Make it Strong Againcited by 1
1 more
  1. 2026STORM: End-to-End Referring Multi-Object Tracking in Videoscited by 1

Built on by

Not cited yet.