Plug-and-Play Acceleration of Occupancy Grid-based NeRF Rendering using VDB Grid and Hierarchical Ray Traversal
This work provides a plug-and-play acceleration method for NeRF rendering, addressing a specific bottleneck in occupancy grid-based approaches for computer graphics and 3D reconstruction applications, but it is incremental as it builds on existing OG techniques.
The paper tackled the inefficiency of Occupancy Grid (OG) in NeRF rendering due to redundant voxel examinations, by replacing dense grids with VDB grids and using hierarchical ray traversal, resulting in average rendering speedups of 12% on NeRF-Synthetic and 4% on Mip-NeRF 360 datasets without quality loss.
Transmittance estimators such as Occupancy Grid (OG) can accelerate the training and rendering of Neural Radiance Field (NeRF) by predicting important samples that contributes much to the generated image. However, OG manages occupied regions in the form of the dense binary grid, in which there are many blocks with the same values that cause redundant examination of voxels' emptiness in ray-tracing. In our work, we introduce two techniques to improve the efficiency of ray-tracing in trained OG without fine-tuning. First, we replace the dense grids with VDB grids to reduce the spatial redundancy. Second, we use hierarchical digital differential analyzer (HDDA) to efficiently trace voxels in the VDB grids. Our experiments on NeRF-Synthetic and Mip-NeRF 360 datasets show that our proposed method successfully accelerates rendering NeRF-Synthetic dataset by 12% in average and Mip-NeRF 360 dataset by 4% in average, compared to a fast implementation of OG, NerfAcc, without losing the quality of rendered images.