Instant Neural Graphics Primitives with a Multiresolution Hash Encoding
Reviewed by Pithpith:5ZW7XOXUopen to challenge →
read the original abstract
Neural graphics primitives, parameterized by fully connected neural networks, can be costly to train and evaluate. We reduce this cost with a versatile new input encoding that permits the use of a smaller network without sacrificing quality, thus significantly reducing the number of floating point and memory access operations: a small neural network is augmented by a multiresolution hash table of trainable feature vectors whose values are optimized through stochastic gradient descent. The multiresolution structure allows the network to disambiguate hash collisions, making for a simple architecture that is trivial to parallelize on modern GPUs. We leverage this parallelism by implementing the whole system using fully-fused CUDA kernels with a focus on minimizing wasted bandwidth and compute operations. We achieve a combined speedup of several orders of magnitude, enabling training of high-quality neural graphics primitives in a matter of seconds, and rendering in tens of milliseconds at a resolution of ${1920\!\times\!1080}$.
This paper has not been read by Pith yet.
Forward citations
Cited by 5 Pith papers
-
Lighting-Consistent Object Transfer Across Radiance Fields
Diffusion-based per-view harmonization for lighting-consistent object transfer between 3DGS scenes, using heterogeneous training data and final 3D consolidation.
-
ReAge3D: Re-Aging 3D Faces with View Consistency
ReAge3D trains a diffusion re-aging model on synthetic pairs then uses masked propagation from a frontal pivot view to produce consistent multi-view images that supervise 3D face optimization.
-
LagrangianSplats: Divergence-Free Transport of Gaussian Primitives for Fluid Reconstruction
A framework that structurally enforces divergence-free velocity and long-range transport coherence in 3D fluid reconstruction from 2D videos via divergence-free kernels advecting Lagrangian Gaussian splats.
-
PREF: Phasorial Embedding Fields for Compact Neural Representations
PREF introduces a phasor volume and tailored Fourier mapping to let shallow MLPs capture high-frequency signals compactly in 2D images, 3D SDFs, and 5D NeRFs.
-
Efficiently Linking Real Scenes with Synthetic Data Generation for AI-based Cognitive Robotics and Computer Vision Applications
The paper reviews limits in AI vision for robotics and describes work-in-progress on bridging sim-to-real domain gaps by linking real and synthetic training data.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.