Reliability-Aware Prediction via Uncertainty Learning for Person Image Retrieval

Abstract

Current person image retrieval methods have achieved great improvements in accuracy metrics. However, they rarely describe the reliability of the prediction. In this paper, we propose an Uncertainty-Aware Learning (UAL) method to remedy this issue. UAL aims at providing reliability-aware predictions by considering data uncertainty and model uncertainty simultaneously. Data uncertainty captures the "noise" inherent in the sample, while model uncertainty depicts the model's confidence in the sample's prediction. Specifically, in UAL, (1) we propose a sampling-free data uncertainty learning method to adaptively assign weights to different samples during training, down-weighting the lowquality ambiguous samples. (2) we leverage the Bayesian framework to model the model uncertainty by assuming the parameters of the network follow a Bernoulli distribution. (3) the data uncertainty and the model uncertainty are jointly learned in a unified network, and they serve as two fundamental criteria for the reliability assessment: if a probe is highquality (low data uncertainty) and the model is confident in the prediction of the probe (low model uncertainty), the final ranking will be assessed as reliable. Experiments under the risk-controlled settings and the multi-query settings show the proposed reliability assessment is effective. Our method also shows superior performance on three challenging benchmarks under the vanilla single query settings. The code is available at: https://github.com/dcp15/UAL

Proxy-based embedding alignment for RGB-infrared person re-identification

“…The key challenge in this task is the appearance discrepancy caused by variations in illumination, background, pose, and viewpoint. The current methods can be roughly divided into two groups: hand-craft methods [15][16][17] and deep-learning methods [2][3][4][5][6][7]18] . Presently, the latter has dominated the re-ID research community.…”

“…Person re-IDentification (re-ID) is a technique for retrieving images of a specific person from a large database [1] , which has numerous applications in various fields, including surveillance, criminal investigations, and smart retail. Although re-ID for the visible image (RGB) scheme has made encouraging progress [2][3][4][5][6][7] , several other pending problems emerge as prerequisites for realistic re-ID systems. Among these problems, a critical one is the RGB-Infrared (RGB-IR) re-ID [8][9][10][11] , which requires matching a person across two different modalities, i.e., the RGB modality and the infrared (IR) modality.…”