Search papers, labs, and topics across Lattice.
The paper introduces 8D Neural Assets (8DNA), a novel neural representation that pre-bakes complex light transport effects, including subsurface scattering and interreflections, into 3D assets. 8DNA learns the full 8D light transport function, enabling accurate rendering even under near-field illumination, which is a limitation of previous 6D approaches. By employing a distribution-learning formulation during training, 8DNA reduces optimization variance and requires a lower training budget compared to regression-based methods, while achieving comparable accuracy to path tracing with faster inference.
Near-field lighting? No problem: 8DNA pre-bakes complex light transport into neural representations, outperforming prior methods with faster inference and lower training costs.
High-fidelity 3D assets exhibit intriguing global illumination effects like subsurface scattering, glossy interreflections, and fine-scale fiber scatterings, which often involve long scattering paths that are expensive to simulate. We introduce 8D neural assets (8DNA) to pre-bake these light transport effects into neural representations. Unlike prior methods that assume far-field lighting and precompute light transport into 6D functions, 8DNA learns the full 8D light transport, enabling accurate rendering under near-field illumination. Our training leverages a distribution-learning formulation that learns light transport from forward path-traced samples, which produces less optimization variance with lower training budget than the prior regression-based approaches. Experiments show our 8DNA rendering closely matches path-traced results under various scene configurations, yet it achieves improved variance reduction and fast inference speeds on challenging assets.
NVIDIA