BVH Encoding for Neural-Assisted Ray Tracing Daniel Meister, Takahiro Harada (AMD - Advanced Micro Devices, Inc.) jcgt.org/published/0015/02/0… Abstract: We propose a novel input encoding method of 3D positions for neural networks used in ray tracing algorithms. Our encoding technique utilizes the bounding volume hierarchy (BVH), which serves both as an acceleration data structure for ray tracing and as a spatial data structure for indexing learnable latent feature vectors. Our method naturally adapts to scene geometry, supports general geometric primitives, and maintains a bounded memory footprint independent of scene complexity. We demonstrate an application of our technique in the context of neural radiance cache (NRC).

Variational r-Adaptive Cloth Simulation Jiahao Wen, Zhen Chen, Jernej Barbič, Danny M. Kaufman (Adobe; USC) arxiv.org/abs/2608.17833 Abstract: We present the first r-adaptive method for simulating cloth dynamics and statics with frictional contact in modern cloth pipelines. Thin cloth requires high effective spatial resolution to reproduce wrinkles, folds, buckling, and sharp contact features. However, applying existing variational r-adaptivity to piecewise-linear shells reveals two coupled failure modes. Discretized incremental-potential (IP) optimization can become trapped in poor local minima, yielding suboptimal physical configurations. It can also lower IP artificially by collapsing elements, invalidating the finite-element approximation on which the objective relies. We address both problems with degeneracy-activated quality regularization. The regularizer remains inactive for well-shaped elements, preserving anisotropic adaptation and local densification, but becomes strong near degeneracy. It suppresses spurious low-energy basins, improves escape from suboptimal physical minima, and prevents element bunching, a cloth-specific failure in which elements progressively collapse as cloth slides across sharp contact features. For practical performance, we introduce a dynamic nonlinear solver that exploits within-timestep coherence through accelerated derivative evaluation and dynamic IPC tolerance updates for r-adaptive iterative trust-region (ITR) solves. This yields a 3-6x speedup over prior optimal ITR. Experiments on challenging frictional-contact scenarios show that, under equal vertex-count and time-budget constraints, our method achieves higher visual fidelity than fixed meshes.

Curly Hair Simulation using Curly Finite Elements Xinming Pei, Zhendong Wang, Lei Lan, Weiwei Xu, Yin Yang, Huamin Wang arxiv.org/abs/2607.22103 Abstract: Realistic simulation of curly hair is challenging due to the tight coupling between macroscopic strand deformation and high-frequency geometric details such as waves and helices. In this paper, we propose a curly hair model that decomposes each strand into curly elements, consisting of a rod-base configuration and an analytically defined high-frequency wrinkles represented by planar waves or volumetric helices, with deformation governed primarily by bending for wavy hair and twisting for spiral hair. A curvature-energy splitting scheme separates stretching, buckling, and bending contributions, and efficient energy approximations improve numerical robustness and reduce computational cost without degrading visual fidelity. We further introduce a hybrid collision handling strategy that combines coarse collision proxies for the base configuration with analytical treatment of high-frequency details, and adapt guide-strand interpolation to our curly finite-element representation to generate dense hair while preserving fine structure. Experiments demonstrate stable, efficient, and visually faithful simulation of curly hair across diverse scenarios.