FM-ReID: Selective Competitive Token Routing for Object Re-Identification

2026-09-28

AI Summary

This study addresses the issue in object re-identification where global similarity obscures locally discriminative features by proposing an end-to-end fine-grained representation learning framework. The core innovation lies in introducing a novel multi-query competitive routing mechanism that dynamically allocates DINOv3 dense tokens via residual slots, enabling the competitive mining of local features without fixed spatial partitioning and thereby effectively enhancing the global descriptive capacity of vision foundation models. Extensive experiments demonstrate that the proposed method achieves significant performance improvements across diverse re-identification benchmarks, including animals, pedestrians, and vehicles, validating both the generalizability and effectiveness of the competitive routing strategy.

Abstract

Object re-identification (ReID) faces a recurring challenge: different identities can share highly similar global appearances, while the cues that distinguish them are localized, heterogeneous, and visible only under particular viewpoints. This challenge arises in animal ReID through markings, contours, and scars, in person ReID through subtle clothing and accessory cues, and in vehicle ReID through localized appearance details. Although visual foundation models encode such information in dense tokens, a single holistic descriptor can obscure discriminative local signals. We propose FM-ReID, an end-to-end framework that formulates local representation learning as selective competitive token routing. Its Competitive Fine-grained Mining module uses multiple mining queries and a residual query to compete for dense DINOv3 tokens. Above-prior selection retains tokens preferentially allocated to each mining query, while the residual slot receives tokens excluded from the retrieval descriptors. The resulting multi-query descriptors are jointly trained with a holistic representation for retrieval, without fixed spatial partitions or equal-area constraints. FM-ReID achieves strong results on animal, person, and vehicle ReID benchmarks, supporting competitive token routing as an effective way to augment holistic foundation-model representations.

Problem

• Object Re-Identification
• Visual Foundation Models
• Local Discriminative Cues
• Holistic Representation

Innovation

• Object Re-Identification
• Competitive Token Routing
• Foundation Model
• Fine-grained Mining
• DINOv3