A5107219809- Lexicography and Language Studies
- Historical Linguistics and Language Studies
- Translation Studies and Practices
- Linguistic Variation and Morphology
- Metaheuristic Optimization Algorithms Research
- Gender Studies in Language
- Human Pose and Action Recognition
- Evolutionary Algorithms and Applications
- Video Surveillance and Tracking Methods
- Industrial Technology and Control Systems
- Advanced Machining and Optimization Techniques
- Natural Language Processing Techniques
- Spanish Linguistics and Language Studies
- Linguistic research and analysis
- Linguistics, Language Diversity, and Identity
- Advanced Decision-Making Techniques
- Swearing, Euphemism, Multilingualism
- Advanced Multi-Objective Optimization Algorithms
- Advanced Sensor and Control Systems
- Smart Grid and Power Systems
- Manufacturing Process and Optimization
- Hand Gesture Recognition Systems
- Fuzzy Logic and Control Systems
- Advanced Surface Polishing Techniques
- Advanced Vision and Imaging
Zhejiang University
2022-2024
Alibaba Group (United States)
2024
Huawei Technologies (China)
2020-2024
Alibaba Group (Cayman Islands)
2023
City University of Macau
2022
University of Macau
2022
Beijing Advanced Sciences and Innovation Center
2016
Beijing University of Technology
2016
Shanghai University
2014
Fudan University
2004-2013
Video-based 3D human pose and shape estimations are evaluated by intra-frame accuracy inter-frame smoothness. Although these two metrics responsible for different ranges of temporal consistency, existing state-of-the-art methods treat them as a unified problem use monotonous modeling structures (e.g., RNN or attention-based block) to design their networks. However, using single kind structure is difficult balance the learning short-term long-term correlations, may bias network one them,...
Generating photorealistic 3D faces from given conditions is a challenging task. Existing methods often rely on time-consuming one-by-one optimization approaches, which are not efficient for modeling the same distribution content, e.g., faces. Additionally, an ideal controllable face generation model should consider both facial attributes and expressions. Thus we propose novel approach called TEx-Face(TExt & Expression-to-Face) that addresses these challenges by dividing task into three...
The goal of sign language recognition (SLR) is to help those who are hard hearing or deaf overcome the communication barrier. Most existing approaches can be typically divided into two lines, i.e., Skeleton-based, and RGB-based methods, but both lines methods have their limitations. Skeleton-based do not consider facial expressions, while usually ignore fine-grained hand structure. To limitations, we propose a new framework called Spatial-temporal Part-aware network (StepNet), based on RGB...
Due to the low power feature of Bluetooth, network based on Bluetooth has been considered as one promising positioning technology candidates by research and industry communities. In order reduce error being introduced Received Signal Strength Indication (RSSI) algorithm, an optimized RSSI ranging algorithm is proposed, which fully technology. First, this uses trilateral get rough region estimation covers all unknown nodes. Second, breaking down area, forms value for beacon nodes at centroid...
The goal of sign language recognition (SLR) is to help those who are hard hearing or deaf overcome the communication barrier. Most existing approaches can be typically divided into two lines, i.e., Skeleton-based and RGB-based methods, but both lines methods have their limitations. do not consider facial expressions, while usually ignore fine-grained hand structure. To limitations, we propose a new framework called Spatial-temporal Part-aware network~(StepNet), based on RGB parts. As its...
IBio-PSE is an efficient and interactive Problem Solving Environment platform (PSE) for Biomolecular Networks Visualization based on website. For facilitating users to analyze various biomolecular networks different comparison result with the web browser, a uniform PSE website built. In this paper, provides extended, researchers visualize interact their network conveniently efficiently. Two applications have been done in end indicate that model suitable visualization visualization.
Video-based 3D human pose and shape estimations are evaluated by intra-frame accuracy inter-frame smoothness. Although these two metrics responsible for different ranges of temporal consistency, existing state-of-the-art methods treat them as a unified problem use monotonous modeling structures (e.g., RNN or attention-based block) to design their networks. However, using single kind structure is difficult balance the learning short-term long-term correlations, may bias network one them,...
Generating photorealistic 3D faces from given conditions is a challenging task. Existing methods often rely on time-consuming one-by-one optimization approaches, which are not efficient for modeling the same distribution content, e.g., faces. Additionally, an ideal controllable face generation model should consider both facial attributes and expressions. Thus we propose novel approach called TEx-Face(TExt & Expression-to-Face) that addresses these challenges by dividing task into three...