---
license: apache-2.0
arxiv: 2505.01257
---

# CAMELTrack
## Context-Aware Multi-cue ExpLoitation for Online Multi-Object Tracking

[![arXiv](https://img.shields.io/badge/arXiv-2505.01257-<COLOR>.svg)](https://arxiv.org/abs/2505.01257) 
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/cameltrack-context-aware-multi-cue-1/multi-object-tracking-on-dancetrack)](https://paperswithcode.com/sota/multi-object-tracking-on-dancetrack?p=cameltrack-context-aware-multi-cue-1)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/cameltrack-context-aware-multi-cue-1/multi-object-tracking-on-sportsmot)](https://paperswithcode.com/sota/multi-object-tracking-on-sportsmot?p=cameltrack-context-aware-multi-cue-1)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/cameltrack-context-aware-multi-cue-1/multi-object-tracking-on-mot17)](https://paperswithcode.com/sota/multi-object-tracking-on-mot17?p=cameltrack-context-aware-multi-cue-1)
<!---
Add PoseTrack21 & BEE24
--->

>**[CAMELTrack: Context-Aware Multi-cue ExpLoitation for Online Multi-Object Tracking](https://arxiv.org/abs/2505.01257)**
>
>Vladimir Somers, Baptiste Standaert, Victor Joos, Alexandre Alahi, Christophe De Vleeschouwer
>
>[*arxiv 2505.01257*](https://arxiv.org/abs/2505.01257)

**CAMELTrack** is an **Online Multi-Object Tracker** that learns to associate detections without hand-crafted heuristics. 
It combines multiple tracking cues through a lightweight, fully trainable module and achieves state-of-the-art performance while 
staying modular and fast.

![](https://github.com/user-attachments/assets/706a6b5a-10f5-4464-97bd-266e737ffcc3)

## 📄 Abstract
**Online Multi-Object Tracking** has been recently dominated by **Tracking-by-Detection** (TbD) methods, where recent advances 
rely on increasingly sophisticated heuristics for tracklet representation, feature fusion, and multi-stage matching. 
The key strength of TbD lies in its modular design, enabling the integration of specialized off-the-shelf models like 
motion predictors and re-identification. However, the extensive usage of human-crafted rules for temporal associations 
makes these methods inherently limited in their ability to capture the complex interplay between various tracking cues. 
In this work, we introduce **CAMEL**, a novel association module for Context-Aware Multi-Cue ExpLoitation, that learns 
resilient association strategies directly from data, breaking free from hand-crafted heuristics while maintaining TbD's 
valuable modularity.