A5015279307- Computer Graphics and Visualization Techniques
- 3D Shape Modeling and Analysis
- Advanced Vision and Imaging
- Evacuation and Crowd Dynamics
- Human Motion and Animation
- Face recognition and analysis
- Traffic control and management
- Advanced Numerical Analysis Techniques
- Generative Adversarial Networks and Image Synthesis
- Membrane Separation Technologies
- Advanced Image and Video Retrieval Techniques
- Adsorption and Cooling Systems
- Opinion Dynamics and Social Influence
- Computational Geometry and Mesh Generation
- Image Processing and 3D Reconstruction
- Chemical Looping and Thermochemical Processes
- Advanced Image Processing Techniques
- Complex Network Analysis Techniques
- Human Pose and Action Recognition
- Membrane-based Ion Separation Techniques
- Image Enhancement Techniques
- Traffic Prediction and Management Techniques
- Autonomous Vehicle Technology and Safety
- Simulation and Modeling Applications
- Visual Attention and Saliency Detection
Zhejiang University
2016-2025
East China University of Science and Technology
2023-2025
Nanjing Tech University
2016-2024
State Key Laboratory of Chemical Engineering
2023-2024
Zhejiang University of Science and Technology
2016-2024
Tencent (China)
2020-2023
Bournemouth University
2023
Hangzhou Academy of Agricultural Sciences
2020
National Renewable Energy Laboratory
2014
Zhejiang University of Technology
2014
In the same vein of discriminative one-shot learning, Siamese networks allow recognizing an object from a single exemplar with class label. However, they do not take advantage underlying structure data and relationship among multitude samples as only rely on pairs instances for training. this paper, we propose new quadruplet deep network to examine potential connections training instances, aiming achieve more powerful representation. We design shared four branches that receive multi-tuple...
We analyze the passengers' traffic pattern for 1.58 million taxi trips of Shanghai, China. By employing non-negative matrix factorization and optimization methods, we find that, people travel on workdays mainly three purposes: commuting between home workplace, traveling from workplace to others such as leisure activities. Therefore, flow in one area or any pair locations can be approximated by a linear combination basis flows, corresponding purposes respectively. name coefficients powers,...
Rapid biomimetic intrafibrillar mineralization of collagen was induced by polyelectrolyte–calcium complexes as a pre-precursor.
Real-time in-between motion generation is universally required in games and highly desirable existing animation pipelines. Its core challenge lies the need to satisfy three critical conditions simultaneously: quality, controllability speed , which renders any methods that offline computation (or post-processing) or cannot incorporate (often unpredictable) user control undesirable. To this end, we propose a new real-time transition method address aforementioned challenges. Our approach...
We present a learning algorithm that uses bone-driven motion networks to predict the deformation of loose-fitting garment meshes at interactive rates. Given garment, we generate simulation database and extract virtual bones from simulated mesh sequences using skin decomposition. At runtime, separately compute low- high-frequency deformations in sequential manner. The low-frequency are predicted by transferring body motions bones' motions, estimated leveraging global information local...
In this paper, we propose a framework of maximizing quadratic submodular energy with knapsack constraint approximately, to solve certain computer vision problems. The proposed maximization problem can be viewed as generalization the classic 0/1 problem. Importantly, our constrained function solved via dynamic programing. We further introduce range-reduction step prior programing two-stage procedure for more efficient maximization. order demonstrate effectiveness and its algorithm, apply it...
Abstract Retinal vessel segmentation is of great significance for assisting doctors in diagnosis ophthalmological diseases such as diabetic retinopathy, macular degeneration and glaucoma. This article proposes a new retinal algorithm using generative adversarial learning with large receptive field. A network maps an input fundus image to realistic while discriminative differentiates between images drawn from the database network. Firstly, proposed combines shallow features upsampled deep...
Abstract Simulation of real‐world traffic scenarios is widely needed in virtual environments. Different from many previous works on simulating vehicles or pedestrians separately, our approach aims to capture the realistic process vehicle–pedestrian interaction for mixed simulation. We model a decision‐making their based gap acceptance judging criterion and then design novel environmental feedback mechanism both vehicles' pedestrians' behavior‐control models drive motions. demonstrate that...
Interactive multi-agent simulation algorithms are used to compute the trajectories and behaviors of different entities in virtual reality scenarios. However, current methods involve considerable parameter tweaking generate plausible behaviors. We introduce a novel approach (Heter-Sim) that combines physics-based with data-driven techniques using an optimization-based formulation. Our is general can simulate heterogeneous agents corresponding human crowds, traffic, vehicles, or combinations...
It is important and urgent to overcome the intermittent nature of solar energy as a green substitute for fossil-based electricity. Concentrated power plants with thermochemical storage are considered potential option cost-effective electricity generation dispatchability. This study aims propose novel concentrated plant that uses based on calcium looping flexible operation sequence eliminate dependence carbonation reaction under sunshine. The involves two successive phases: charging in...