Robust Duality Learning for Unsupervised Visible-Infrared Person Re-Identification

计算机科学 无监督学习 鉴定(生物学) 对偶(序理论) 人工智能 红外线的 模式识别(心理学) 机器学习 数学 光学 物理 植物 离散数学 生物
作者
Yongxiang Li,Yuan Sun,Yang Qin,Dezhong Peng,Xi Peng,Peng Hu
出处
期刊:IEEE Transactions on Information Forensics and Security [Institute of Electrical and Electronics Engineers]
卷期号:20: 1937-1948 被引量:28
标识
DOI:10.1109/tifs.2025.3536613
摘要

Unsupervised visible-infrared person re-identification (UVI-ReID) aims at retrieving pedestrian images of the same individual across distinct modalities, presenting challenges due to the inherent heterogeneity gap and the absence of cost-prohibitive annotations. Although existing methods employ self-training with clustering-generated pseudo-labels to bridge this gap, they always implicitly assume that these pseudo-labels are predicted correctly. In practice, however, this presumption is impossible to satisfy due to the difficulty of training a perfect model let alone without any ground truths, resulting in pseudo-labeling errors. Based on the observation, this study introduces a new learning paradigm for UVI-ReID considering Pseudo-Label Noise (PLN), which encompasses three challenges: noise overfitting, error accumulation, and noisy cluster correspondence. To conquer these challenges, we propose a novel robust duality learning framework (RoDE) for UVI-ReID to mitigate the adverse impact of noisy pseudo-labels. Specifically, for noise overfitting, we propose a novel Robust Adaptive Learning mechanism (RAL) to dynamically prioritize clean samples while deprioritizing noisy ones, thus avoiding overemphasizing noise. To circumvent error accumulation of self-training, where the model tends to confirm its mistakes, RoDE alternately trains dual distinct models using pseudo-labels predicted by their counterparts, thereby maintaining diversity and avoiding collapse into noise. However, this will lead to cross-cluster misalignment between the two distinct models, not to mention the misalignment between different modalities, resulting in dual noisy cluster correspondence and thus difficult to optimize. To address this issue, a Cluster Consistency Matching mechanism (CCM) is presented to ensure reliable alignment across distinct modalities as well as across different models by leveraging cross-cluster similarities. Extensive experiments on three benchmark datasets demonstrate the effectiveness of the proposed RoDE.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
机智臻发布了新的文献求助10
刚刚
刚刚
思源的应助被冷静的柜子采纳,获得10
1秒前
EASA发布了新的文献求助10
1秒前
xxxthe发布了新的文献求助10
1秒前
Akim的应助被2Q采纳,获得10
2秒前
风趣的荠发布了新的文献求助10
3秒前
铭坤完成签到 ,获得积分10
3秒前
zd200572完成签到,获得积分10
4秒前
秋风的应助被深情的猫咪采纳,获得10
4秒前
4秒前
4秒前
酷波er的应助被想人陪的蚂蚁采纳,获得10
4秒前
李健的小迷弟的应助被1282941496采纳,获得10
4秒前
4秒前
6秒前
6秒前
Ava的应助被无限觅风采纳,获得10
6秒前
7秒前
yu123发布了新的文献求助10
7秒前
忧伤的老四完成签到,获得积分10
7秒前
8秒前
DJC完成签到,获得积分10
9秒前
9秒前
11发布了新的文献求助10
9秒前
烟花的应助被许自通采纳,获得10
10秒前
10秒前
10秒前
暖暖发布了新的文献求助10
10秒前
11秒前
thomas发布了新的文献求助10
11秒前
12秒前
山海不远发布了新的文献求助10
12秒前
chiaugustus完成签到,获得积分10
12秒前
15秒前
梦梦的应助被东十八采纳,获得10
15秒前
15秒前
桃桃奶盖发布了新的文献求助10
15秒前
abydauw发布了新的文献求助10
16秒前
白塔发布了新的文献求助20
16秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
plantTFDB 508
Resource Recovery Handbook 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8175336
求助须知:如何正确求助?哪些是违规求助? 9567158
关于积分的说明 21790491
捐赠科研通 7769866
什么是DOI,文献DOI怎么找? 3431849
关于科研通互助平台的介绍 2556178
邀请新用户注册赠送积分活动 2477280