---
title: "MoTRa: Motion-Aware Target Representation Learning for End-to-End Multi-Object Tracking"
authors:
  - "Yuanzhou Huang"
  - "Songwei Pei"
  - "Shuhuai Wang"
  - "Bingfeng Liu"
  - "Qian Li"
  - "Shangguang Wang"
doi: "10.24963/ijcai.2026/137"
url: "https://ijcai.org/proceedings/2026/137"
pdf: "https://ijcai.org/proceedings/2026/0137.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "1224-1232"
track: "Main Track"
keywords:
  - "Computer Vision: Video analysis and understanding"
  - "Computer Vision: Motion and tracking"
  - "Computer Vision: Representation learning"
---

# MoTRa: Motion-Aware Target Representation Learning for End-to-End Multi-Object Tracking

Yuanzhou Huang, Songwei Pei, Shuhuai Wang, Bingfeng Liu, Qian Li, Shangguang Wang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 1224-1232. DOI: [10.24963/ijcai.2026/137](https://doi.org/10.24963/ijcai.2026/137)

## Abstract

Multi-object tracking (MOT) has long faced challenges with identity switches, especially for targets with low appearance discriminability and complex motion. Existing end-to-end trackers typically enhance robustness by modeling long-term temporal information across target-level representations, yet these representations remain insufficiently discriminative for spatially and visually similar targets. In this paper, we present MoTRa, a Motion-aware Target Representation learning framework for end-to-end MOT that enriches target-level representations with adaptive motion cues. As a result, each representation adaptively balances motion cues and appearance-related content cues under varying conditions, enhancing its discriminability for tracking. To achieve this, we propose a feature fusion and alignment module that extracts motion cues and adaptively fuses them with content features into target representations. To further regularize the dynamically fused representations, we propose a target-specific contrastive learning strategy that promotes intra-trajectory consistency and inter-target separability. Experimental results demonstrate the effectiveness of our method, achieving competitive performance on multiple benchmarks.