Yubo Zhang

ORCID: 0000-0001-5539-0200
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Computer Graphics and Visualization Techniques
  • 3D Shape Modeling and Analysis
  • Advanced Vision and Imaging
  • Distributed Control Multi-Agent Systems
  • Computational Geometry and Mesh Generation
  • Human Motion and Animation
  • Geological Modeling and Analysis
  • Mobile Crowdsensing and Crowdsourcing
  • Neural Networks Stability and Synchronization
  • Digital Media and Visual Art
  • Fluid Dynamics and Heat Transfer
  • Control and Stability of Dynamical Systems
  • Linguistics and language evolution
  • Diffusion and Search Dynamics
  • Image and Video Quality Assessment
  • Quantum and electron transport phenomena
  • Color perception and design
  • Topological Materials and Phenomena
  • Autonomous Vehicle Technology and Safety
  • Creativity in Education and Neuroscience
  • Automated Road and Building Extraction
  • Linguistic Variation and Morphology
  • Video Surveillance and Tracking Methods
  • Facility Location and Emergency Management
  • Rare-earth and actinide compounds

Zhengzhou University
2022-2024

Henan University of Engineering
2024

Aerospace Information Research Institute
2024

Chinese Academy of Sciences
2024

Huawei Technologies (China)
2023

Nvidia (United States)
2018-2019

University of California, Davis
2012-2017

Wuhan University
2016

Princeton Plasma Physics Laboratory
2015

Inner Mongolia University
2014

High quality global illumination can enhance the visual perception of depth cue and local thickness volumetric data but it is seldom used in scientific visualization because its high computational cost. This paper presents a novel grid-based technique which specially designed optimized for volume purpose. It supports common light sources dynamic transfer function editing. Our method models propagation, including both absorption scattering, using convection-diffusion equation that be solved...

10.1145/2448196.2448205 article EN 2013-03-21

Rendering techniques that increase realism in volume visualization help enhance perception of the 3D features data. While focusing on high-quality global illumination have been extensively studied, few works handle interaction light with materials volume. Existing for light-material are limited their ability to high-frequency real-world material data, and current treatment data poorly supports correct integration surface materials. In this paper, we introduce an alternative definition...

10.1109/tvcg.2017.2784830 article EN publisher-specific-oa IEEE Transactions on Visualization and Computer Graphics 2017-12-19

We present a technique that adaptively adjusts the shading rate during rendering based on scene content and motion. Our goal is to improve performance with no loss in perceived quality. determine per-screen-tile by testing an error estimate against perceptually-corrected just-noticeable difference threshold. design features effective efficient using spatial frequency analysis of half quarter shading. also study effect motion reducing error, consequence display-persistence and/or blur...

10.1145/3320287 article EN Proceedings of the ACM on Computer Graphics and Interactive Techniques 2019-06-03

We introduce a novel spatio-temporal extrapolation technique for fluid simulation designed to improve the results without using higher resolution grids. In general, there are rigid demands associated with pushing animations resolutions given limited computational capabilities. This in tradeoffs between implementing high-order numerical methods and increasing of space time. For 3D problems, such challenges rapidly become cost-ineffective. The method we present improves flow features this...

10.1145/2508363.2508401 article EN ACM Transactions on Graphics 2013-11-01

Traditional computational fluid dynamics (CFD) solvers are usually written for a single gridding paradigm such as structured-Cartesian, structured-body-fitted, or unstructured grids. Each type of mesh paradigms has inherent advantages and disadvantages. Thus, the methods coupling multiple have been developed to facilitate use different in part domain. However, complex hybrid poses challenges rendering calculations visualizing data. This paper describes volume visualization system...

10.1109/ldav.2014.7013209 article EN 2014-11-01

We introduce a new method to efficiently track complex interfaces among multi-phase immiscible fluids. Unlike existing techniques, we use mesh-based representation for global liquid surfaces while selectively modeling some local surficial regions with regional level sets (RLS) handle geometries that are difficult resolve explicit topology operations. Such semi-explicit surface mechanism can preserve volume, fine features and foam-like thin films under relatively low computational...

10.1109/tvcg.2018.2864283 article EN IEEE Transactions on Visualization and Computer Graphics 2018-08-07

This article presents a joint study between computer scientists and fusion in developing visual tools for studying patterns flow fields from large-scale magnetic confinement simulations. The authors visualize time-varying data by generating trajectory curves via massless particle advection design set of color functions, pathline filters, projection methods specific to achieve research objectives. These aid managing the complexity large datasets are crucial locating understanding subtle...

10.1109/mcse.2015.107 article EN Computing in Science & Engineering 2015-09-23

The benefits of using advanced illumination models in volume visualization have been demonstrated by many researchers. Interactive rendering incorporated with lighting has achieved GPU acceleration for regular-grid data, making even more appealing as a tool 3D data exploration. This paper presents an interactive strategy, which is specially designed and optimized unstructured-grid data. basis the design partial differential equation based model to simulate light propagation, absorption,...

10.1109/pacificvis.2015.7156383 article EN 2015-04-01

Simulating a lively forest in real time is challenging problem due to the high complexities both scene geometry and dynamic effects. In this paper, we introduce novel hybrid model efficiently synthesize realistic swaying effects of trees under winds. An approximate lighting algorithm also proposed render lakes. Our method effectively takes advantage horse power modern GPUs. Nearly all simulations are done GPU little data transfer needed in-between CPU GPU. We reach real-time frame rates...

10.1109/apccas.2008.4746098 article EN 2008-11-01

We present a new real-time shadow rendering approach whose kernel is hierarchical disk-based approximation to the scene geometry. show that this approximated representation greatly accelerates visibility inquiry, facilitating computation of anisotropic occlusion in GPU. By taking account occlusion, our can simulate cast from not only ambient light, but also point, directional and environment lights. Because no pre-computation required, it suitable for dynamically deforming objects. 1. Motivation

10.2312/egs.20061024 article EN Eurographics 2006-01-01

In this poster, we present a volume rendering framework that achieves realtime of global illumination effects for datasets, such as multiple scattering and shadow. This approach incorporates the volumetric photon mapping technique [Jensen Christensen 1998] into classical precomputed radiance transfer [Sloan et al. 2002] pipeline. Fig.1 shows our method is successfully applied in both interactive graphics scientific visualization applications.

10.1145/2159616.2159663 article EN 2012-03-09

We provide nearly optimal algorithms for online facility location (OFL) with predictions. In OFL, $n$ demand points arrive in order and the algorithm must irrevocably assign each point to an open upon its arrival. The objective is minimize total connection costs from assigned facilities plus opening cost. further assume additionally given $x_i$ a natural prediction $f_{x_i}^{\mathrm{pred}}$ which supposed be $f_{x_i}^{\mathrm{opt}}$ that serves offline solution. Our main result $O(\min\{\log...

10.48550/arxiv.2110.08840 preprint EN cc-by arXiv (Cornell University) 2021-01-01

In the multi-agent system, uniqueness of formation shape is a difficult point in research. This paper proposes distributed algorithm system to predefined without using global information. proposed algorithm, each node holds local coordinate common orientation, and has only distance measurement information between its neighbors. Because no needed, restriction on configuration agent reduced. By barycentric coordinates with respect neighbors, can achieve desired shape. Simulation results have...

10.1109/yac57282.2022.10023834 article EN 2022 37th Youth Academic Annual Conference of Chinese Association of Automation (YAC) 2022-11-19
Coming Soon ...