</> Code with Papers by Syntology

Papers › Relation Preserving Triplet Mining for Stabilising the Triplet Loss in...

Relation Preserving Triplet Mining for Stabilising the Triplet Loss in Re-identification Systems

15 Oct 2021Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision 2023 1arXiv:2110.07933archive 2025-07-28

Adhiraj Ghosh, Kuruparan Shanmugalingam, Wen-Yan Lin

Object appearances change dramatically with pose variations. This creates a challenge for embedding schemes that seek to map instances with the same object ID to locations that are as close as possible. This issue becomes significantly heightened in complex computer vision tasks such as re-identification(reID). In this paper, we suggest that these dramatic appearance changes are indications that an object ID is composed of multiple natural groups, and it is counterproductive to forcefully map instances from different groups to a common location. This leads us to introduce Relation Preserving Triplet Mining (RPTM), a feature-matching guided triplet mining scheme, that ensures that triplets will respect the natural subgroupings within an object ID. We use this triplet mining mechanism to establish a pose-aware, well-conditioned triplet loss by implicitly enforcing view consistency. This allows a single network to be trained with fixed parameters across datasets while providing state-of-the-art results. Code is available at https://github.com/adhirajghosh/RPTM_reid.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

adhirajghosh/rptm_reid officialmentioned in papermentioned on GitHubpytorchMIT 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

ObjectPerson Re-IdentificationVehicle Re-Identification

2 archive task tags without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Person Re-Identification DukeMTMC-reID RPTM Rank-1 93.5 #12 of 94 Archive leaderboard report
Person Re-Identification DukeMTMC-reID RPTM Rank-5 96.1 #12 of 94 Archive leaderboard report
Person Re-Identification DukeMTMC-reID RPTM mAP 89.2 #12 of 94 Archive leaderboard report
Vehicle Re-Identification VeRi-776 RPTM Rank-1 97.3 #2 of 17 Archive leaderboard report
Vehicle Re-Identification VeRi-776 RPTM Rank1 97.3 #2 of 17 Archive leaderboard report
Vehicle Re-Identification VeRi-776 RPTM Rank5 98.4 #2 of 17 Archive leaderboard report
Vehicle Re-Identification VeRi-776 RPTM mAP 88.0 #2 of 17 Archive leaderboard report
Vehicle Re-Identification VehicleID Large RPTM Rank-1 92.9 #4 of 10 Archive leaderboard report
Vehicle Re-Identification VehicleID Large RPTM Rank-5 96.3 #4 of 10 Archive leaderboard report
Vehicle Re-Identification VehicleID Large RPTM mAP 80.5 #4 of 10 Archive leaderboard report
Vehicle Re-Identification VehicleID Medium RPTM Rank-1 93.3 #4 of 9 Archive leaderboard report
Vehicle Re-Identification VehicleID Medium RPTM Rank-5 96.5 #4 of 9 Archive leaderboard report
Vehicle Re-Identification VehicleID Medium RPTM mAP 81.2 #4 of 9 Archive leaderboard report
Vehicle Re-Identification VehicleID Small RPTM Rank-1 95.5 #4 of 13 Archive leaderboard report
Vehicle Re-Identification VehicleID Small RPTM Rank-5 97.4 #4 of 13 Archive leaderboard report
Vehicle Re-Identification VehicleID Small RPTM mAP 84.8 #4 of 13 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