A5101368879- Video Surveillance and Tracking Methods
- Human Pose and Action Recognition
- Multimodal Machine Learning Applications
- Face recognition and analysis
- Gait Recognition and Analysis
- Pharmacovigilance and Adverse Drug Reactions
- Hand Gesture Recognition Systems
- Domain Adaptation and Few-Shot Learning
- Photoacoustic and Ultrasonic Imaging
- Lipoproteins and Cardiovascular Health
- Cancer-related molecular mechanisms research
- Spondyloarthritis Studies and Treatments
- Pelvic and Acetabular Injuries
- Robotics and Sensor-Based Localization
- Scoliosis diagnosis and treatment
- Epilepsy research and treatment
- Advanced Neural Network Applications
- Treatment of Major Depression
- Sexual function and dysfunction studies
- Soft Robotics and Applications
Zhejiang University
2023-2025
Shenzhen University
2024
Xiamen University
2024
Recently, person re-identification (ReID) has witnessed fast development due to its broad practical applications and proposed various settings, e.g., traditional ReID, clothes-changing visible-infrared ReID. However, current studies primarily focus on single specific tasks, which limits model applicability in real-world scenarios. This paper aims address this issue by introducing a novel instruct-ReID task that unifies 6 existing ReID tasks one retrieves images based provided visual or...
Human-centric perception tasks, e.g., pedestrian detection, skeleton-based action recognition, and pose estimation, have wide industrial applications, such as metaverse sports analysis. There is a recent surge to develop human-centric foundation models that can benefit broad range of tasks. While many achieved success, they did not explore 3D vision-language tasks for required task-specific finetuning. These limitations restrict their application more downstream situations. To tackle these...
Background: Diazepam, one of the benzodiazepines, is widely used clinically to treat anxiety, for termination epilepsy, and sedation. However, reports its adverse events (AEs) have been numerous, even fatal complications reported. In this study, we investigated AEs diazepam based on real data from U.S. Food Drug Administration (FDA) event reporting system (FAERS). Methods: Disproportionality in diazepam-associated was assessed through calculation odds ratios (RORs), proportional (PRRs),...
Background: Atorvastatin is a commonly prescribed medication for the prevention of cardiovascular diseases. Recent observational studies have suggested potential association between atorvastatin use and occurrence Erectile Dysfunction (ED). In this study, we aimed to explore relationship ED using real-world data from FAERS database employed Mendelian randomization assess causality. Methods: To evaluate disproportionality in relation ED, conducted several pharmacovigilance analyses, including...
Human-centric perception tasks, e.g., pedestrian detection, skeleton-based action recognition, and pose estimation, have wide industrial applications, such as metaverse sports analysis. There is a recent surge to develop human-centric foundation models that can benefit broad range of tasks. While many achieved success, they did not explore 3D vision-language tasks for required task-specific finetuning. These limitations restrict their application more downstream situations. To tackle these...
Grounded language-image pre-trained models have shown strong zero-shot generalization to various downstream object detection tasks. Despite their promising performance, the rely heavily on laborious prompt engineering. Existing works typically address this problem by tuning text prompts using training data in a few-shot or fully supervised manner. However, rarely studied is optimize without any annotations. In paper, we delve into and propose an Unsupervised Prompt Tuning framework for...
Human intelligence can retrieve any person according to both visual and language descriptions. However, the current computer vision community studies specific re-identification (ReID) tasks in different scenarios separately, which limits applications real world. This paper strives resolve this problem by proposing a novel instruct-ReID task that requires model images given image or instructions. Instruct-ReID is first exploration of general ReID setting, where existing 6 be viewed as special...
Abstract In deep neural networks, performance can degrade when test data distributions differ from training data. Unsupervised Domain Generalization (UDG) aims to improve generalization across unseen domains by leveraging multiple source without supervision. Traditional methods focus on extracting domain‐invariant features, potentially at the expense of feature space integrity and potential. We presents a Multi‐Domain Representation Network (MDRN) for unsupervised multi‐domain learning. MDRN...
Human intelligence can retrieve any person according to both visual and language descriptions. However, the current computer vision community studies specific re-identification (ReID) tasks in different scenarios separately, which limits applications real world. This paper strives resolve this problem by proposing a new instruct-ReID task that requires model images given image or instructions. Our is more general ReID setting, where existing 6 be viewed as special cases designing We propose...