Papers › When to Extract ReID Features: A Selective Approach for Improved Multiple Object Tracking

When to Extract ReID Features: A Selective Approach for Improved Multiple Object Tracking

10 Sep 2024arXiv:2409.06617archive 2025-07-28

Emirhan Bayar, Cemal Aker

Extracting and matching Re-Identification (ReID) features is used by many state-of-the-art (SOTA) Multiple Object Tracking (MOT) methods, particularly effective against frequent and long-term occlusions. While end-to-end object detection and tracking have been the main focus of recent research, they have yet to outperform traditional methods in benchmarks like MOT17 and MOT20. Thus, from an application standpoint, methods with separate detection and embedding remain the best option for accuracy, modularity, and ease of implementation, though they are impractical for edge devices due to the overhead involved. In this paper, we investigate a selective approach to minimize the overhead of feature extraction while preserving accuracy, modularity, and ease of implementation. This approach can be integrated into various SOTA methods. We demonstrate its effectiveness by applying it to StrongSORT and Deep OC-SORT. Experiments on MOT17, MOT20, and DanceTrack datasets show that our mechanism retains the advantages of feature extraction during occlusions while significantly reducing runtime. Additionally, it improves accuracy by preventing confusion in the feature-matching stage, particularly in cases of deformation and appearance similarity, which are common in DanceTrack. https://github.com/emirhanbayar/Fast-StrongSORT, https://github.com/emirhanbayar/Fast-Deep-OC-SORT

PaperPDFCode

Code

emirhanbayar/fast-deep-oc-sort officialmentioned in papermentioned on GitHubpytorch report
emirhanbayar/fast-strongsort 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 TrackingMultiple Object TrackingObject DetectionObject Trackingobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Multi-Object Tracking DanceTrack Fast-StrongSORT AssA 38.8 #27 of 37 Archive leaderboard report
Multi-Object Tracking DanceTrack Fast-StrongSORT HOTA 55.9 #27 of 37 Archive leaderboard report
Multi-Object Tracking DanceTrack Fast-StrongSORT IDF1 54.6 #27 of 37 Archive leaderboard report
Multi-Object Tracking MOT17 Fast-StrongSORT AssA 62.3 #17 of 48 Archive leaderboard report
Multi-Object Tracking MOT17 Fast-StrongSORT HOTA 62.7 #17 of 48 Archive leaderboard report
Multi-Object Tracking MOT17 Fast-StrongSORT IDF1 77.5 #17 of 48 Archive leaderboard report
Multi-Object Tracking MOT20 Fast-StrongSORT HOTA 61.2 #14 of 27 Archive leaderboard report
Multi-Object Tracking MOT20 Fast-StrongSORT IDF1 75.4 #14 of 27 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.

Methods

Focus

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