A5078774034- Video Surveillance and Tracking Methods
- Human Pose and Action Recognition
- Gait Recognition and Analysis
- Advanced Neural Network Applications
Hangzhou First People's Hospital
2025
Westlake University
2025
Nanjing University of Information Science and Technology
2022-2023
Unsupervised person re-identification (Re-ID) aims at finding the most informative features from unlabeled datasets. Some recent approaches adopted camera-aware strategies for model training and have thereby achieved highly promising results. However, these methods simultaneously address intra-ID discrepancies of all cameras require independent learning under each camera, which increases complexity algorithm. To resolve this issue, we present a camera contrast framework unsupervised Re-ID....
Person re-identification (re-ID) aims to match the same person across different cameras. However, most existing re-ID methods assume that people wear clothes in views, which limit their performance identifying target pedestrians who change clothes. Cloth-changing is a quite challenging problem as occupying large number of pixels an image becomes invalid or even misleads information. To tackle this problem, we propose novel Multi-biometric Unified Network (MBUNet) for learning robustness...
The goal of unsupervised person re-identification (Re-ID) is to use unlabeled images learn discriminative features. In recent years, many approaches have adopted clustered pseudo labels construct proxies for contrastive learning, and thereby achieved great success. However, existing methods this kind only utilize local structures within IDs design their while ignoring the relations between samples different IDs, which limits improvement inter-ID ability. To resolve issue, we propose a Global...
Cardiac arrest (CA) poses a significant global health challenge and often results in poor prognosis. We developed an interpretable applicable machine learning (ML) model for predicting in-hospital mortality of CA patients who survived more than 72 h. A total 721 were extracted from the Medical Information Mart Intensive Care IV database, divided into training set (n = 576) internal validation 145). The external containing 856 cases collected four tertiary hospitals Zhejiang Province. primary...