Ray Tracing of Signed Distance Function Grids

We evaluate the performance of a wide set of combinations of traversal and voxel intersection testing of signed distance function grids in a path tracing setting. In addition, we present an optimized way to compute the intersection between a ray and the surface defined by trilinear interpolation of signed distances at the eight corners of a voxel. We also provide a novel way to compute continuous normals across voxels and an optimization for shadow rays.

Learning Robust Real-World Dexterous Grasping Policies via Implicit Shape Augmentation

Noah Brenowitz

Climate Simulation

Climate Simulation

Shape, Light, and Material Decomposition from Images using Monte Carlo Rendering and Denoising

Abstract



Accelerated Policy Learning with Parallel Differentiable Simulation

Deep reinforcement learning can generate complex control policies, but requires large amounts of training data to work effectively. Recent work has attempted to address this issue by leveraging differentiable simulators. However, inherent problems such as local minima and exploding/vanishing numerical gradients prevent these methods from being generally applied to control tasks with complex contact-rich dynamics, such as humanoid locomotion in classical RL benchmarks.

Jie Xu

Jie Xu is a Research Scientist in Seattle Robotics Lab, NVIDIA Research. His research mainly focuses on the intersection of Robotics, Machine Learning, and Computer Graphics.  Prior to NVIDIA, he received his Ph.D. degree in Computer Science in 2022 at MIT CSAIL in the Computational Design and Fabrication Group (CDFG) and obtained his bachelor's degree from the Department of Computer Science and Technology at Tsinghua University with honors in 2016. 

Yue Wang

My research lies in the intersection of computer vision, computer graphics, and robotics. My goal is to use machine learning to enable robot intelligence with minimal human supervision. I study how to design 3D learning systems which leverage geometry, appearance, and any other cues that are naturally available in sensory inputs. I am also broadly interested in eclectic applications on top of these systems. More info can be found in my website.

Dennis Abts

Dennis has three decades of experience building large-scale parallel computers that are uniquely capable of tackling the most demanding AI and HPC workloads. Previously, as the Chief Architect at Groq he worked on large-scale parallel architectures for machine learning, and at Google he worked on warehouse-scale topologies for energy-proportional networking, and Cray, where he was a Sr.

Apoorva Sharma

Apoorva Sharma is a Research Scientist in the Autonomous Vehicles Group at NVIDIA Research. His research interests focus on quantifying uncertainty in machine learning, with application towards building safe ML-enabled autonomous systems.

Subscribe to