A5087452083- Computer Graphics and Visualization Techniques
- 3D Shape Modeling and Analysis
- Human Motion and Animation
- 3D Surveying and Cultural Heritage
- Video Analysis and Summarization
- Face recognition and analysis
- Image Processing and 3D Reconstruction
- Advanced Vision and Imaging
- Human Pose and Action Recognition
- COVID-19 epidemiological studies
- SARS-CoV-2 and COVID-19 Research
- Augmented Reality Applications
- Facial Nerve Paralysis Treatment and Research
- Geological Modeling and Analysis
- Innovative Educational Techniques
- Satellite Image Processing and Photogrammetry
- Generative Adversarial Networks and Image Synthesis
- Viral gastroenteritis research and epidemiology
- Viral Infections and Outbreaks Research
- Image Enhancement Techniques
- Heavy Metal Exposure and Toxicity
- Textile materials and evaluations
- Hepatitis C virus research
- Respiratory viral infections research
- Traffic and Road Safety
Zhejiang University
2003-2024
PLA Army Engineering University
2024
Zhejiang Lab
2023-2024
Tencent (China)
2022-2023
National Administration of Surveying, Mapping and Geoinformation of China
2017
Air Force Medical University
2011
Union Hospital
2004
Beijing Red Cross Blood Center
2004
Chinese Academy of Sciences
2003
BGI Group (China)
2003
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...
The widespread threat of severe acute respiratory syndrome (SARS) to human life has spawned challenges develop fast and accurate analytical methods for its early diagnosis create a safe antiviral vaccine preventive use. Consequently, we thoroughly investigated the immunoreactivities with patient sera series synthesized peptides from SARS-coronavirus structural proteins.We 41 ranging in size 16 25 amino acid residues relatively high hydrophilicity. SARS were determined by ELISA.Four epitopic...
Styled online in-between motion generation has important application scenarios in computer animation and games. Its core challenge lies the need to satisfy four critical requirements simultaneously: speed, quality, style diversity, synthesis controllability. While first two challenges demand a delicate balance between simple fast models learning capacity for latter are rarely investigated together existing methods, which largely focus on either control without or uncontrolled stylized...
To understand the time-course of viraemia and antibody responses to severe acute respiratory syndrome-associated coronavirus (SARS-CoV), RT-PCR ELISA were used assay 376 blood samples from 135 SARS patients at various stages illness, including who in their early convalescent phase. The results showed that IgM antibodies decreased became undetectable 11 weeks into recovery IgG antibodies, however, remained detectable for a period beyond found 100 % SARS-CoV mainly appeared 1 week after onset...
Existing neural rendering-based text-to-3D-portrait generation methods typically make use of human geometry prior and diffusion models to obtain guidance. However, relying solely on information introduces issues such as the Janus problem, over-saturation, over-smoothing. We present Portrait3D , a novel framework with joint geometry-appearance achieve that overcomes aforementioned issues. To accomplish this, we train 3D portrait generator, 3DPortraitGAN, robust prior. This generator is...
We present deep shapely portraits, a novel method based on learning, to automatically reshape an input portrait be better proportioned and more while keeping personal facial characteristics. Different from existing methods that may suffer irrational face artifacts when dealing with portraits large pose variations or reshaping adjustments, we utilize dense 3D information constraints instead of sparse landmarks morphable models, resulting in reshaped faces lying rational space. To this end,...
Woven slit-film geotextiles are often subjected to in-plane tensile loads in engineering applications, which may alter relevant permeability properties. The fractal model of the coefficient woven is extended predict uniaxial and laterally constrained strains. Based on observation summary variation pore size distribution pattern with strain, unit introduced. expressed as functions dimension, characteristics, physical parameters weft strain. A clamping device capable applying tension invented....
Existing neural rendering-based text-to-3D-portrait generation methods typically make use of human geometry prior and diffusion models to obtain guidance. However, relying solely on information introduces issues such as the Janus problem, over-saturation, over-smoothing. We present Portrait3D, a novel framework with joint geometry-appearance achieve that overcomes aforementioned issues. To accomplish this, we train 3D portrait generator, 3DPortraitGAN-Pyramid, robust prior. This generator is...
During their service life, needle-punched nonwoven geotextiles are frequently subjected to uniaxial tensile strains, which may impact permeability behavior. The basic model of the coefficient in was extended strain condition, considering influence fiber orientation distribution. is expressed as a function tortuosity fractal dimension, pore size characteristics, orientation, physical parameters, and strain. To address errors results traditional tests attributed inhomogeneity multiple layers...
Motion style transfer changes the of a motion while retaining its content and is useful in computer animations games. Contact an essential component that should be controlled explicitly order to express vividly enhancing naturalness quality. However, it unknown how decouple control contact achieve fine-grained transfer. In this paper, we present novel method for over contacts achieving both spatial-temporal variations style. Based on our empirical evidence, propose controlling indirectly...
Style-guided texture generation aims to generate a that is harmonious with both the style of reference image and geometry input mesh, given 3D mesh its text description. Although diffusion-based methods, such as distillation sampling, have numerous promising applications in stylized games films, it requires addressing two challenges: 1) decouple content completely from for models, 2) align generated color tone, image, prompt. To this end, we introduce StyleTex, an innovative...
Style-guided texture generation aims to generate a that is harmonious with both the style of reference image and geometry input mesh, given 3D mesh its text description. Although diffusion-based methods, such as distillation sampling, have numerous promising applications in stylized games films, it requires addressing two challenges: 1) decouple content completely from for models, 2) align generated color tone, image, prompt. To this end, we introduce StyleTex, an innovative...
Abstract A large body of evidence indicates that abnormalities in the levels iron, copper and zinc their metabolism are associated with neurodegenerative diseases. However, it is difficult to decide whether any observed changes trace elements reflect primary disease process or secondary a mechanism. In present study, Fe, Cu Zn organs transgenic mice which express familial Alzheimer's (AD) gene normal same species ages were determined by X‐ray fluorescence (XRF) spectrometry. The results show...
3D-aware face generators are typically trained on 2D real-life image datasets that primarily consist of near-frontal data, and as such, they unable to construct one-quarter headshot 3D portraits with complete head, neck, shoulder geometry. Two reasons account for this issue: First, existing facial recognition methods struggle extracting data captured from large camera angles or back views. Second, it is challenging learn a distribution covering the region single-view due significant...
Sharing short personalized videos to various social media networks has become quite popular in recent years. This raises the need for digital retouching of portraits videos. However, applying portrait image editing directly on video frames cannot generate smooth and stable sequences. To this end, we present a robust easy-to-use parametric method reshape produce retouched results. Given an input video, our consists two main stages: stabilized face reconstruction, continuous reshaping. In...
SurfaceBrush and Brush2Model are two systems which enable users to create 3D objects intuitively using a hand-held controller in virtual reality (VR). These state-of-the-art methods either start modeling from dense collections of stroke ribbons drawn by professional artists, or the most basic point skeletons, line polygon skeletons. Thus, it is very challenging for novices amateurs design complex models efficiently. We propose 3D-BrushVR, novel VR tool that uses volume skeleton-based...
Abstract. GF-7 satellite is a two-line-array stereo imaging for surveying and mapping which will be launched in 2018. Its resolution about 0.8 meter at subastral point corresponding to 20 km width of cloth, the viewing angle its forward backward cameras are 5 26 degrees. This paper proposed simulation method images. WorldView-2 images were used as basic data simulation. That is, we didn’t use DSM DOM (we call it “ortho-to-stereo” method) but “stereo-to-stereo” method, better reflect...
Abstract In this paper we present a novel shape‐constrained fireworks simulation method with rich textures in an HMD (Helmet Mounted Display) virtual environment using sketched feature lines as input. Our approach first retrieves object from three‐dimensional (3D) model database sketch‐based 3D shape retrieval algorithm. Then, order to approximate models complex structures, introduce point sampling algorithm based on Gaussian curvatures, which stores not only the positions of selected...