A5017967708- Computer Graphics and Visualization Techniques
- 3D Shape Modeling and Analysis
- Advanced Vision and Imaging
- Welding Techniques and Residual Stresses
- Robotics and Automated Systems
- Wireless Body Area Networks
- Biometric Identification and Security
- Face recognition and analysis
- Hydrogen embrittlement and corrosion behaviors in metals
- Color perception and design
- Product Development and Customization
- Medical Image Segmentation Techniques
- Cellular Automata and Applications
- Advanced Machining and Optimization Techniques
- Advanced Welding Techniques Analysis
- Manufacturing Process and Optimization
- Facial Rejuvenation and Surgery Techniques
- Advanced machining processes and optimization
- Facial Nerve Paralysis Treatment and Research
- Generative Adversarial Networks and Image Synthesis
University of Science and Technology of China
2014-2024
Hangzhou Dianzi University
2022-2023
We present a novel semantic model for human head defined with neural radiance field. The 3D-consistent consist of set disentangled and interpretable bases, can be driven by low-dimensional expression coefficients. Thanks to the powerful representation ability field, constructed represent complex facial attributes including hair, wearings, which not represented traditional mesh blendshape. To construct personalized model, we propose define bases as several multi-level voxel fields. With short...
We introduce D$^3$-Human, a method for reconstructing Dynamic Disentangled Digital Human geometry from monocular videos. Past video human reconstruction primarily focuses on undecoupled clothed bodies or only clothing, making it difficult to apply directly in applications such as animation production. The challenge decoupled clothing and body lies the occlusion caused by over body. To this end, details of visible area plausibility invisible must be ensured during process. Our proposed...
Various SDF-based neural implicit surface reconstruction methods have been proposed recently, and demonstrated remarkable modeling capabilities. However, due to the global nature limited representation ability of a single network, existing still suffer from many drawbacks, such as accuracy scale reconstruction. In this paper, we propose versatile, scalable high-quality address these issues. We integrate divide-and-conquer approach into Specifically, model object or scene fusion multiple...
The welding and construction processes for H-type thick-plate bridge steel involve complex multi-pass processes, which make it difficult to ensure its performance. Accordingly, is crucial explore the inherent correlations between process parameters quality, apply them robots, eliminating instability in manual welding. In order improve GMAW (gas metal arc welding) are simulated, using Q345qD flat joint model. Four welds with X-shaped grooves designed optimize of current, voltage, speed....
In this paper, we propose a virtual 3D Eyeglasses Try-on (3DET) system, with efficient, realistic and real-time augmented performance. The 3DET system captures user's performance the help of depth camera renders glasses properly on video stream immediately after selection. eye-glasses follow motion (movement or rotation) head simultaneously. high efficiency our is achieved by simplifying eyeglasses matching procedure, making use active appearance model (AAM) based face tracking algorithm....
The welding and construction processes of H-type thick-plate-bridge steel involve complex multi-pass processes, which is difficulty to ensure its performance. Accordingly, it crucial explore the inherent correlation among parameters quality, applied into robots, eliminating instability from manual welding. In order improve GMAW (Gas Metal Arc Welding) process are simulated using Q345qD bridge flat joint model. Four welds with X-shaped grooves designed optimize current, voltage, speed....
This paper investigates the transmission of three-dimensional (3D) human face content for immersive communication over a rate-constrained transmitter-receiver link. We propose new framework named NeRF-SeCom, which leverages neural radiance fields (NeRF) and semantic communications to improve quality 3D visualizations while minimizing overhead. In NeRF-SeCom framework, we first train NeRF model based on NeRFBlendShape method, is pre-shared between transmitter receiver as knowledge base...
(1) Background: Personalized product customization is an important direction in the development of chair industry. This paper studies intelligent design method for rapid realization personalized office customization; (2) Methods: based on case-based reasoning (CBR) method, characteristic attributes patterns are analyzed, and attribute model established. According to data customer demand, using multi-layer weighted k-nearest neighbor (K-NN) developed entropy weight analytic hierarchy process....