Beibei Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Beibei Wang has a total of 89 co-authors. Here are the top co-authors by number of shared publications:
- Lu Wang
Shandong University
15 shared publications - Ling-Qi Yan
Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, United Arab Emirates
13 shared publications - Jian Yang
Nanjing University of Science and Technology, Nanjing, China
10 shared publications - Miloš Hašan
Nvidia, San Jose, USA
9 shared publications - Nicolas Holzschuch
Univ. Grenoble-Alpes, Inria, CNRS, Grenoble INP, LJK, 38000 Grenoble, France
9 shared publications - Ligang Liu
University of Science and Technology of China, Hefei, China
5 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (6) , 1-15
Macrofacet Theory for Gaussian Process Statistical Surfaces
We present macrofacet theory to extend microfacet theory from the microspace to the macro-space. This is achieved by transforming surfaces into volumetric representations that preserve microfacet characteristics. Therefore, we formulate a macroscopic microfacet model using a classic exponential participating medium. Meanwhile, we observe that traditional microfacet models are equivalent to Gaussian processes by definition but ignore the correlation along the geometric normal of the macro-surface. We extend microfacet theory to address this limitation. Our formulation represents Gaussian process implicit surfaces in a statistical manner, which we refer to as Gaussian process statistical surfaces. As a result, our approach converts Gaussian process statistical surfaces into classic exponential media to render surfaces, volumes and in-betweens without realizations. This enables efficient rendering and improves performance compared to realization-based approaches, while theoretically bridging microfacet models and Gaussian processes. Moreover, our approach is easy to implement.
- 9 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (6) , 1-15
Open AccessMacrofacet Theory for Gaussian Process Statistical Surfaces
We present macrofacet theory to extend microfacet theory from the microspace to the macro-space. This is achieved by transforming surfaces into volumetric representations that preserve microfacet characteristics. Therefore, we formulate a macroscopic microfacet model using a classic exponential participating medium. Meanwhile, we observe that traditional microfacet models are equivalent to Gaussian processes by definition but ignore the correlation along the geometric normal of the macro-surface. We extend microfacet theory to address this limitation. Our formulation represents Gaussian process implicit surfaces in a statistical manner, which we refer to as Gaussian process statistical surfaces. As a result, our approach converts Gaussian process statistical surfaces into classic exponential media to render surfaces, volumes and in-betweens without realizations. This enables efficient rendering and improves performance compared to realization-based approaches, while theoretically bridging microfacet models and Gaussian processes. Moreover, our approach is easy to implement.
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (6) , 1-17
Differentiable Rendering for Specular Materials via Projected Specular Manifolds
Physics-based differentiable rendering computes derivatives of radiometric measures with respect to scene parameters. A key challenge lies in handling scene discontinuities from geometries, materials, and other factors. These discontinuities lead to a unique boundary integral defined on discontinuity curves. While occlusion-induced discontinuities have been well studied, those caused by materials, especially specular materials, remain largely unexplored. Specular bidirectional scattering distribution functions (BSDFs) contain Dirac delta distributions arising from the laws of perfect reflection or refraction, which constrain light path geometry and restrict the path integral to the specular manifold. The specular manifold is defined by geometric constraints and lacks explicit location, making the evaluation of discontinuity curves challenging. To address this, our key insight is that we can locate the specular manifold without explicit computation by projecting it onto a subregion of a specular object, which we call the projected specular manifold (PSM). After projection, while exact locations of discontinuity curves remain unknown, we can adopt the warped-area reparameterization to convert the boundary integral into an interior integral over the surface, avoiding the need to precisely locate these curves. The key challenge of this reparameterization is constructing an additional velocity field. For this, our key observation is that for a specular path, the distances from vertices to manifold boundaries and associated velocities are known. Therefore, we propose a simple but practical direct-interpolation-based method to construct the velocity. Furthermore, we extend the warped-area sampling to handle visibility boundaries with specular BSDFs and generalize our method to glossy materials. Our computed derivatives match references with minor additional computational cost, and can successfully drive gradient-based optimization tasks like shape and position estimation. Our implementation is available at https://github.com/Rhythm25/PSM.
- 9 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (6) , 1-17
Open AccessDifferentiable Rendering for Specular Materials via Projected Specular Manifolds
Physics-based differentiable rendering computes derivatives of radiometric measures with respect to scene parameters. A key challenge lies in handling scene discontinuities from geometries, materials, and other factors. These discontinuities lead to a unique boundary integral defined on discontinuity curves. While occlusion-induced discontinuities have been well studied, those caused by materials, especially specular materials, remain largely unexplored. Specular bidirectional scattering distribution functions (BSDFs) contain Dirac delta distributions arising from the laws of perfect reflection or refraction, which constrain light path geometry and restrict the path integral to the specular manifold. The specular manifold is defined by geometric constraints and lacks explicit location, making the evaluation of discontinuity curves challenging. To address this, our key insight is that we can locate the specular manifold without explicit computation by projecting it onto a subregion of a specular object, which we call the projected specular manifold (PSM). After projection, while exact locations of discontinuity curves remain unknown, we can adopt the warped-area reparameterization to convert the boundary integral into an interior integral over the surface, avoiding the need to precisely locate these curves. The key challenge of this reparameterization is constructing an additional velocity field. For this, our key observation is that for a specular path, the distances from vertices to manifold boundaries and associated velocities are known. Therefore, we propose a simple but practical direct-interpolation-based method to construct the velocity. Furthermore, we extend the warped-area sampling to handle visibility boundaries with specular BSDFs and generalize our method to glossy materials. Our computed derivatives match references with minor additional computational cost, and can successfully drive gradient-based optimization tasks like shape and position estimation. Our implementation is available at https://github.com/Rhythm25/PSM.
- 1 October 2026
- Published by Springer Nature
- p. 286-303
3D Gaussian Texture for Real-Time Mesoscale Appearance Synthesis and Rendering
- 1 October 2026
- Published by Springer Nature
- p. 286-303
3D Gaussian Texture for Real-Time Mesoscale Appearance Synthesis and Rendering
- Open Access
- 25 August 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (5) , 1
Erratum: Fiber-level Woven Fabric Capture from a Single Microscopic Image
This is an erratum for the article “Fiber-level Woven Fabric Capture from a Single Microscopic Image” published in ACM Trans. Graph. 45, 5, Article 43 (June 2026), 18 pages.
- 25 August 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (5) , 1
Open AccessErratum: Fiber-level Woven Fabric Capture from a Single Microscopic Image
This is an erratum for the article “Fiber-level Woven Fabric Capture from a Single Microscopic Image” published in ACM Trans. Graph. 45, 5, Article 43 (June 2026), 18 pages.
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-11
Relit-LiVE: Relight Video by Jointly Learning Environment Video
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-11
Open AccessRelit-LiVE: Relight Video by Jointly Learning Environment Video
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-11
PureSample: Neural Materials Learned by Sampling Microgeometry
Traditional physically-based material models rely on analytically derived bidirectional reflectance distribution functions (BRDFs), typically by considering statistics of micro-primitives such as facets, flakes, or spheres, sometimes combined with multi-bounce interactions such as layering and multiple scattering. These derivations are often complex and model-specific. Once an analytic BRDF evaluation is defined, one still needs to design an importance sampling method for it and evaluate the probability density function (pdf) of that sampling distribution, requiring further model-specific derivations. We present PureSample: a novel neural BRDF representation that allows learning a material’s appearance purely by sampling forward random walks on the microgeometry, which is usually straightforward to implement. Our representation allows for efficient BRDF evaluation, importance sampling, and pdf evaluation, for homogeneous as well as spatially varying materials. We achieve this by two learnable components: first, the sampling distribution is modeled using a flow matching neural network, which allows both importance sampling and pdf evaluation; second, we introduce a view-dependent albedo term, captured by a lightweight neural network, which allows for converting a pdf value to a BRDF value for any pair of view and light directions. We demonstrate PureSample on challenging materials, including various microgeometries, multi-layered materials, and multiple-scattering microfacet materials.
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-11
Open AccessPureSample: Neural Materials Learned by Sampling Microgeometry
Traditional physically-based material models rely on analytically derived bidirectional reflectance distribution functions (BRDFs), typically by considering statistics of micro-primitives such as facets, flakes, or spheres, sometimes combined with multi-bounce interactions such as layering and multiple scattering. These derivations are often complex and model-specific. Once an analytic BRDF evaluation is defined, one still needs to design an importance sampling method for it and evaluate the probability density function (pdf) of that sampling distribution, requiring further model-specific derivations. We present PureSample: a novel neural BRDF representation that allows learning a material’s appearance purely by sampling forward random walks on the microgeometry, which is usually straightforward to implement. Our representation allows for efficient BRDF evaluation, importance sampling, and pdf evaluation, for homogeneous as well as spatially varying materials. We achieve this by two learnable components: first, the sampling distribution is modeled using a flow matching neural network, which allows both importance sampling and pdf evaluation; second, we introduce a view-dependent albedo term, captured by a lightweight neural network, which allows for converting a pdf value to a BRDF value for any pair of view and light directions. We demonstrate PureSample on challenging materials, including various microgeometries, multi-layered materials, and multiple-scattering microfacet materials.
- Open Access
- 29 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (5) , 1-18
Fiber-level Woven Fabric Capture from a Single Microscopic Image
Accurately rendering the appearance of fabrics is challenging, due to their complex 3D microstructures and specialized optical properties. If we model the geometry and optics of fabrics down to the fiber level, we can achieve unprecedented rendering realism, but this raises the difficulty of authoring the fiber-level assets. Existing approaches can obtain fiber-level geometry with special devices (e.g., CT) or hand-designed procedural pipelines. In this paper, we propose a method to capture fiber-level geometry and appearance of woven fabrics using a single low-cost microscope image. This may seem like an impossible task: a single image from a low-cost microscope looks very different from the final rendering we would like to achieve, and the information contained in it may seem minimal. We propose a novel fiber parameter estimation pipeline in a coarse-to-fine manner, establishing a subset of parameters step by step. At the core of our pipeline are differentiable procedural geometric and appearance models for woven fabrics at the fiber level, enabling both geometry and appearance to be optimized simultaneously. We first use a simple neural network to predict initial parameters, then we optimize the parameters of procedural fiber geometry and an approximated shading model via differentiable rasterization to match the microscope photo more accurately. Finally, we refine the fiber appearance parameters via differentiable path tracing, converging to accurate fiber optical parameters, which are suitable for physically-based light simulations to produce high-quality rendered results. We believe that our method is the first to utilize differentiable rendering at the microscopic level, supporting physically-based scattering from explicit fiber assemblies. Our fabric parameter estimation achieves high-quality re-rendering of measured woven fabric samples in both distant and close-up views. We also propose a patch-space fiber geometry procedural generation method and a two-scale path tracing framework for efficient rendering of fabric scenes.
- 29 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (5) , 1-18
Open AccessFiber-level Woven Fabric Capture from a Single Microscopic Image
Accurately rendering the appearance of fabrics is challenging, due to their complex 3D microstructures and specialized optical properties. If we model the geometry and optics of fabrics down to the fiber level, we can achieve unprecedented rendering realism, but this raises the difficulty of authoring the fiber-level assets. Existing approaches can obtain fiber-level geometry with special devices (e.g., CT) or hand-designed procedural pipelines. In this paper, we propose a method to capture fiber-level geometry and appearance of woven fabrics using a single low-cost microscope image. This may seem like an impossible task: a single image from a low-cost microscope looks very different from the final rendering we would like to achieve, and the information contained in it may seem minimal. We propose a novel fiber parameter estimation pipeline in a coarse-to-fine manner, establishing a subset of parameters step by step. At the core of our pipeline are differentiable procedural geometric and appearance models for woven fabrics at the fiber level, enabling both geometry and appearance to be optimized simultaneously. We first use a simple neural network to predict initial parameters, then we optimize the parameters of procedural fiber geometry and an approximated shading model via differentiable rasterization to match the microscope photo more accurately. Finally, we refine the fiber appearance parameters via differentiable path tracing, converging to accurate fiber optical parameters, which are suitable for physically-based light simulations to produce high-quality rendered results. We believe that our method is the first to utilize differentiable rendering at the microscopic level, supporting physically-based scattering from explicit fiber assemblies. Our fabric parameter estimation achieves high-quality re-rendering of measured woven fabric samples in both distant and close-up views. We also propose a patch-space fiber geometry procedural generation method and a two-scale path tracing framework for efficient rendering of fabric scenes.
- 30 March 2026
- Published by Wiley in Computer Graphics Forum
HiMat: DiT‐based Ultra‐High Resolution SVBRDF Generation
Creating ultra‐high‐resolution spatially varying bidirectional reflectance functions (SVBRDFs) is critical for photorealistic 3D content creation, to faithfully represent fine‐scale surface details required for close‐up rendering. However, achieving 4K generation faces two key challenges: (1) the need to synthesize multiple reflectance maps at full resolution, which multiplies the pixel budget and imposes prohibitive memory and computational cost, and (2) the requirement to maintain strong pixellevel alignment across maps at 4K, which is particularly difficult when adapting pretrained models designed for the RGB image domain. We introduce HiMat, a diffusion‐based framework tailored for efficient and diverse 4K SVBRDF generation. To address the first challenge, HiMat generates in a high‐compression latent space via a DC‐AE and employs a pretrained diffusion transformer with linear attention to improve per‐map efficiency. To address the second challenge, we propose CrossStitch, a lightweight convolutional module that enforces cross‐map consistency without incurring the cost of global attention. Our experiments show that HiMat achieves high‐fidelity 4K SVBRDF generation with superior efficiency, structural consistency, and diversity compared to prior methods. Beyond materials, our framework also generalizes to related applications such as intrinsic decomposition.
- 30 March 2026
- Published by Wiley in Computer Graphics Forum
HiMat: DiT‐based Ultra‐High Resolution SVBRDF Generation
Creating ultra‐high‐resolution spatially varying bidirectional reflectance functions (SVBRDFs) is critical for photorealistic 3D content creation, to faithfully represent fine‐scale surface details required for close‐up rendering. However, achieving 4K generation faces two key challenges: (1) the need to synthesize multiple reflectance maps at full resolution, which multiplies the pixel budget and imposes prohibitive memory and computational cost, and (2) the requirement to maintain strong pixellevel alignment across maps at 4K, which is particularly difficult when adapting pretrained models designed for the RGB image domain. We introduce HiMat, a diffusion‐based framework tailored for efficient and diverse 4K SVBRDF generation. To address the first challenge, HiMat generates in a high‐compression latent space via a DC‐AE and employs a pretrained diffusion transformer with linear attention to improve per‐map efficiency. To address the second challenge, we propose CrossStitch, a lightweight convolutional module that enforces cross‐map consistency without incurring the cost of global attention. Our experiments show that HiMat achieves high‐fidelity 4K SVBRDF generation with superior efficiency, structural consistency, and diversity compared to prior methods. Beyond materials, our framework also generalizes to related applications such as intrinsic decomposition.
- Open Access
- 14 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Sample Space Partitioning and Spatiotemporal Resampling for Specular Manifold Sampling
- 14 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Open AccessSample Space Partitioning and Spatiotemporal Resampling for Specular Manifold Sampling
- Open Access
- 14 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Diffusion-Guided Relighting for Single-Image SVBRDF Estimation
- 14 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Open AccessDiffusion-Guided Relighting for Single-Image SVBRDF Estimation