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

TL;DR

This paper introduces Relation Preserving Triplet Mining (RPTM), a novel triplet mining scheme that respects natural subgroupings within object IDs to improve re-identification performance despite pose variations.

Contribution

It proposes RPTM, a feature-matching guided triplet mining method that maintains natural groupings and enhances pose-aware triplet loss for better re-identification accuracy.

Findings

Achieves state-of-the-art re-ID results across datasets.

Enables training a single network with fixed parameters.

Improves view consistency and embedding stability.

Abstract

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.