Physically-simulated models for human motion can generate high-quality responsive character animations, often in real-time. Natural language serves as a flexible interface for controlling these models, allowing expert and non-expert users to quickly create and edit their animations. Many recent physics-based animation methods, including those that use text interfaces, train control policies using reinforcement learning (RL). However, scaling these methods beyond several hundred motions has remained challenging.

Hair strands can be modeled by sweeping spheres with varying radii along Bézier curves. We ray-trace such shapes by finding intersections of a given ray with a set of capsules dynamically defined at runtime. A substantial performance boost is achieved by systematically eliminating parts of the shape that are guaranteed not to intersect with the given ray. The new intersector is more than twice faster than the previously leading phantom algorithm. This improvement results in a 30% overall performance increase, which includes traversal, shading, and the rendering system overhead.

Text-to-image models offer a new level of creative flexibility by allowing users to guide the image generation process through natural language. However, using these models to consistently portray the same subject across diverse prompts remains challenging. Existing approaches fine-tune the model to teach it new words that describe specific user-provided subjects or add image conditioning to the model. These methods require lengthy per-subject optimization or large-scale pre-training.

Free-space diffractions are an optical phenomenon where light appears to “bend” around the geometric edges and corners of scene objects. In this paper we present an efficient method to simulate such effects. We derive an edge-based formulation of Fraunhofer diffraction, which is well suited to the common (triangular) geometric meshes used in computer graphics. Our method dynamically constructs a free-space diffraction BSDF by considering the geometry around the intersection point of a ray of light with an object, and we present an importance sampling strategy for these BSDFs.

Stochastic geometry models have enjoyed immense success in graphics for modeling interactions of light with complex phenomena such as participating media, rough surfaces, fibers, and more. Although each of these models operates on the same principle of replacing intricate geometry by a random process and deriving the average light transport across all instances thereof, they are each tailored to one specific application and are fundamentally distinct.

We present fVDB, a novel GPU-optimized framework for deep learning on large-scale 3D data. fVDB 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. fVDB 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.

Monte Carlo rendering algorithms often utilize correlations between pixels to improve efficiency and enhance image quality. For real-time applications in particular, repeated reservoir resampling offers a powerful framework to reuse samples both spatially in an image and temporally across multiple frames. While such techniques achieve equal-error up to 100× faster for real-time direct lighting [Bitterli et al. 2020] and global illumination [Ouyang et al. 2021; Lin et al. 2021], they are still far from optimal.

Numerous scientific and engineering applications require solving boundary value problems (BVPs) like the Laplace and Poisson equations on geometrically intricate domains. We describe a unified Monte Carlo approach to solving elliptic BVPs with Dirichlet, Neumann and Robin boundary conditions using the walk on stars algorithm, which unlike conventional numerical methods, does not require any cumbersome finite element mesh generation or global solves.

Recent advancements in spatiotemporal reservoir resampling (ReSTIR) leverage sample reuse from neighbors to efficiently evaluate the path integral.  Like rasterization, ReSTIR methods implicitly assume a pinhole camera and evaluate the light arriving at a pixel through a single predetermined subpixel location at a time (e.g., the pixel center).  This prevents efficient path reuse in and near pixels with high-frequency details.

Emerging spatial computing systems seamlessly superimpose digital information on the physical environment observed by a user, enabling transformative experiences across various domains, such as entertainment, education, communication and training1,2,3.