Instant neural graphics primitives: lightning fast NeRF and more

[MICRO'23, MLSys'22] TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.

Code for KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs

Implementation of the KinectFusion approach in modern C++14 and CUDA

tutorial for writing custom pytorch cpp+cuda kernel, applied on volume rendering (NeRF)

[SIGGRAPH 2021] ROSEFusion is proposed to tackle the difficulties in fast-motion camera tracking using random optimization with depth information only.

[CVPR'21] [Jittor & Pytorch] DI-Fusion: Online Implicit 3D Reconstruction with Deep Priors

Blazingly fast encoding for neural networks based on permutohedral lattices

INS-Conv: Incremental Sparse Convolution for Online 3D Segmentation (CVPR 2022)

Trilinear sampler with smoothstep and double backpropagation

This is the official code repository for OccuSeg, a state-of-the-art method for accurate joint 3D semantic and instance segmentation.