Clement Fuji Tsang
Clement Fuji Tsang has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Maria Shugrina
Nvidia, Toronto, Canada
2 shared publications - Vismay Modi
NVIDIA, Texas, USA
2 shared publications - Anita Hu
Nvidia, Toronto, Canada
1 shared publication - Eftychios Sifakis
NVIDIA, Madison, WI, USA
1 shared publication - Francis Williams
NVIDIA, New York, USA
1 shared publication - Gergely Klar
NVIDIA Research, Wellington, New Zealand
1 shared publication
Latest Publications
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-3
From 3D Captures to Simulated Digital Environments
We introduce a novel capture-to-simulation pipeline for in-the-wild 3D Gaussian Splat scenes, showing our research-based solutions for segmenting objects, predicting their volumetric mechanical properties using a recent learning-based method, and mixed splat-mesh physics simulations, all supported by recent features in the Kaolin Library. In this hands-on lab, attendees will follow along from their laptop, running Python apps and notebooks in the web browser while compute-heavy AI inference and simulation are performed on reserved cloud GPU back ends. By the end of the course, attendees will be able to use our latest tools to select, process, and simulate objects in a 3D captured scene, and will understand the basics of the segmentation, material assignment, and physics simulation methods that make this pipeline possible.
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Open AccessFrom 3D Captures to Simulated Digital Environments
We introduce a novel capture-to-simulation pipeline for in-the-wild 3D Gaussian Splat scenes, showing our research-based solutions for segmenting objects, predicting their volumetric mechanical properties using a recent learning-based method, and mixed splat-mesh physics simulations, all supported by recent features in the Kaolin Library. In this hands-on lab, attendees will follow along from their laptop, running Python apps and notebooks in the web browser while compute-heavy AI inference and simulation are performed on reserved cloud GPU back ends. By the end of the course, attendees will be able to use our latest tools to select, process, and simulate objects in a 3D captured scene, and will understand the basics of the segmentation, material assignment, and physics simulation methods that make this pipeline possible.
- Open Access
- 10 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Tackling Gaussian Splats, Physics Simulation and Visualization with NVIDIA Kaolin and Warp Libraries
- 10 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Open AccessTackling Gaussian Splats, Physics Simulation and Visualization with NVIDIA Kaolin and Warp Libraries
- Open Access
- 27 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Painting with 3D Gaussian Splat Brushes
- 27 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Open AccessPainting with 3D Gaussian Splat Brushes
- Open Access
- 19 July 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 43 (4) , 1-15
fVDB : A Deep-Learning Framework for Sparse, Large Scale, and High Performance Spatial Intelligence
We present f VDB, a novel GPU-optimized framework for deep learning on large-scale 3D data. f VDB provides a complete set of differentiable primitives to build deep learning architectures for common tasks in 3D learning such as convolution, pooling, attention, ray-tracing, meshing, etc. f VDB simultaneously provides a much larger feature set (primitives and operators) than established frameworks with no loss in efficiency: our operators match or exceed the performance of other frameworks with narrower scope. Furthermore, f VDB can process datasets with much larger footprint and spatial resolution than prior works, while providing a competitive memory footprint on small inputs. To achieve this combination of versatility and performance, f VDB relies on a single novel VDB index grid acceleration structure paired with several key innovations including GPU accelerated sparse grid construction, convolution using tensorcores, fast ray tracing kernels using a Hierarchical Digital Differential Analyzer algorithm (HDDA), and jagged tensors. Our framework is fully integrated with PyTorch enabling interoperability with existing pipelines, and we demonstrate its effectiveness on a number of representative tasks such as large-scale point-cloud segmentation, high resolution 3D generative modeling, unbounded scale Neural Radiance Fields, and large-scale point cloud reconstruction.
- 19 July 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 43 (4) , 1-15
Open AccessfVDB : A Deep-Learning Framework for Sparse, Large Scale, and High Performance Spatial Intelligence
We present f VDB, a novel GPU-optimized framework for deep learning on large-scale 3D data. f VDB provides a complete set of differentiable primitives to build deep learning architectures for common tasks in 3D learning such as convolution, pooling, attention, ray-tracing, meshing, etc. f VDB simultaneously provides a much larger feature set (primitives and operators) than established frameworks with no loss in efficiency: our operators match or exceed the performance of other frameworks with narrower scope. Furthermore, f VDB can process datasets with much larger footprint and spatial resolution than prior works, while providing a competitive memory footprint on small inputs. To achieve this combination of versatility and performance, f VDB relies on a single novel VDB index grid acceleration structure paired with several key innovations including GPU accelerated sparse grid construction, convolution using tensorcores, fast ray tracing kernels using a Hierarchical Digital Differential Analyzer algorithm (HDDA), and jagged tensors. Our framework is fully integrated with PyTorch enabling interoperability with existing pipelines, and we demonstrate its effectiveness on a number of representative tasks such as large-scale point-cloud segmentation, high resolution 3D generative modeling, unbounded scale Neural Radiance Fields, and large-scale point cloud reconstruction.