From Calibration to Refinement: Seeking Certainty via Probabilistic Evidence Propagation for Noisy-Label Person Re-Identification

2026-02-26

AI Summary

This work addresses the challenge of person re-identification under unconstrained settings with severe label noise and sparse samples per identity, where existing methods are prone to overfitting erroneous labels and discarding hard positive samples. To tackle this, the authors propose CARE, a two-stage framework: in the calibration stage, Probabilistic Evidence Calibration (PEC) mitigates the overconfidence of Softmax; in the refinement stage, Evidence Propagation (EPR), Composite Angular Margin (CAM), and Certainty-Oriented Spherical Weighting (COSW) jointly operate in a hyperspherical embedding space to accurately separate clean from noisy samples while preserving hard positives. Extensive experiments demonstrate that CARE achieves state-of-the-art performance on Market1501, DukeMTMC-ReID, and CUHK03 under both random and structured label noise.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Computer Vision: Bias, Fairness & Privacy

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

With the increasing demand for robust person Re-ID in unconstrained environments, learning from datasets with noisy labels and sparse per-identity samples remains a critical challenge. Existing noise-robust person Re-ID methods primarily rely on loss-correction or sample-selection strategies using softmax outputs. However, these methods suffer from two key limitations: 1) Softmax exhibits translation invariance, leading to over-confident and unreliable predictions on corrupted labels. 2) Conventional sample selection based on small-loss criteria often discards valuable hard positives that are crucial for learning discriminative features. To overcome these issues, we propose the CAlibration-to-REfinement (CARE) method, a two-stage framework that seeks certainty through probabilistic evidence propagation from calibration to refinement. In the calibration stage, we propose the probabilistic evidence calibration (PEC) that dismantles softmax translation invariance by injecting adaptive learnable parameters into the similarity function, and employs an evidential calibration loss to mitigate overconfidence on mislabeled samples. In the refinement stage, we design the evidence propagation refinement (EPR) that can more accurately distinguish between clean and noisy samples. Specifically, the EPR contains two steps: Firstly, the composite angular margin (CAM) metric is proposed to precisely distinguish clean but hard-to-learn positive samples from mislabeled ones in a hyperspherical space; Secondly, the certainty-oriented sphere weighting (COSW) is developed to dynamically allocate the importance of samples according to CAM, ensuring clean instances drive model updates. Extensive experimental results on Market1501, DukeMTMC-ReID, and CUHK03 datasets under both random and patterned noises show that CARE achieves competitive performance.

Problem

• noisy-label learning
• person re-identification
• softmax overconfidence
• sample selection
• hard positive mining

Innovation

• noisy-label learning
• person re-identification
• probabilistic evidence propagation
• evidential calibration
• angular margin