Papers › CAMELTrack: Context-Aware Multi-cue ExpLoitation for Online Multi-Object Tracking

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

2 May 2025arXiv:2505.01257archive 2025-07-28

Vladimir Somers, Baptiste Standaert, Victor Joos, Alexandre Alahi, Christophe De Vleeschouwer

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. At its core, CAMEL employs two transformer-based modules and relies on a novel association-centric training scheme to effectively model the complex interactions between tracked targets and their various association cues. Unlike end-to-end detection-by-tracking approaches, our method remains lightweight and fast to train while being able to leverage external off-the-shelf models. Our proposed online tracking pipeline, CAMELTrack, achieves state-of-the-art performance on multiple tracking benchmarks. Our code is available at https://github.com/TrackingLaboratory/CAMELTrack.

PaperPDFCode

Code

TrackingLaboratory/CAMELTrack officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Multi-Object TrackingObject TrackingOnline Multi-Object Tracking

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Multi-Object Tracking DanceTrack CAMELTrack (fully online) HOTA 69.3 #7 of 37 Archive leaderboard report
Multi-Object Tracking MOT17 CAMELTrack (fully online) AssA 61.4 #19 of 48 Archive leaderboard report
Multi-Object Tracking MOT17 CAMELTrack (fully online) DetA 63.6 #19 of 48 Archive leaderboard report
Multi-Object Tracking MOT17 CAMELTrack (fully online) HOTA 62.4 #19 of 48 Archive leaderboard report
Multi-Object Tracking MOT17 CAMELTrack (fully online) IDF1 63.6 #19 of 48 Archive leaderboard report
Multi-Object Tracking MOT17 CAMELTrack (fully online) MOTA 78.5 #19 of 48 Archive leaderboard report
Multi-Object Tracking SportsMOT CAMELTrack (fully online) AssA 72.8 #3 of 22 Archive leaderboard report
Multi-Object Tracking SportsMOT CAMELTrack (fully online) DetA 88.8 #3 of 22 Archive leaderboard report
Multi-Object Tracking SportsMOT CAMELTrack (fully online) HOTA 80.4 #3 of 22 Archive leaderboard report
Multi-Object Tracking SportsMOT CAMELTrack (fully online) IDF1 84.8 #3 of 22 Archive leaderboard report
Multi-Object Tracking SportsMOT CAMELTrack (fully online) MOTA 96.3 #3 of 22 Archive leaderboard report
Online Multi-Object Tracking SportsMOT CAMELTrack HOTA 80.4 #1 of 1 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections