Awesome

NeXT

This repository contains the code release for NeXT: Towards High Quality Neural Radiance Fields via Multi-Skip Transformer. This implementation is based on JaxNeRF and Mip-NeRF. NeXT can be easily incorporated into NeRF and Mip-NeRF methods to serve as the query network, achieving higher quality rendering results.

Abstract

Neural Radiance Fields (NeRF) methods show impressive performance for novel view synthesis by representing a scene via a neural network. However, most existing NeRF based methods, including its variants, treat each sample point individually as input, while ignoring the inherent relationships between adjacent sample points from the corresponding rays, thus hindering the reconstruction performance. To address this issue, we explore a brand new scheme, namely NeXT, introducing a multi-skip transformer to capture the rich relationships between various sample points in a ray-level query. Specifically, ray tokenization is proposed to represent each ray as a sequence of point embeddings which is taken as input of our proposed NeXT. In this way, relationships between sample points are captured via the built-in self-attention mechanism to promote the reconstruction. Besides, our proposed NeXT can be easily combined with other NeRF based methods to improve their rendering quality. Extensive experiments conducted on three datasets demonstrate that NeXT significantly outperforms all previous state-of-the-art work by a large margin. In particular, the proposed NeXT surpasses the strong NeRF baseline by 2.74 PSNR on Blender dataset.

Installation

Following JaxNeRF.

We recommend using Anaconda to set up the environment. Run the following commands:

# Clone the repo
svn export https://github.com/google-research/google-research/trunk/jaxnerf
# Create a conda environment, note you can use python 3.6-3.8 as
# one of the dependencies (TensorFlow) hasn't supported python 3.9 yet.
conda create --name jaxnerf python=3.6.12; conda activate jaxnerf
# Prepare pip
conda install pip; pip install --upgrade pip
# Install requirements
pip install -r jaxnerf/requirements.txt
# [Optional] Install GPU and TPU support for Jax
# Remember to change cuda101 to your CUDA version, e.g. cuda110 for CUDA 11.0.
pip install --upgrade jax jaxlib==0.1.57+cuda101 -f https://storage.googleapis.com/jax-releases/jax_releases.html

Then, you'll need to download the datasets from the NeRF official Google Drive. Please download the nerf_synthetic.zip and unzip to jaxnerf/data and mipnerf/data.

Generate multiscale dataset

You can generate the multiscale dataset by running the following command,

cd mipnerf
python scripts/convert_blender_data.py --blenderdir ./data/nerf_synthetic --outdir ./data/nerf_synthetic_multiscale

Running

Example scripts for training NeXT on Blender and Multi-scale Bender dataset can be found in jaxnerf/scripts and mipnerf/scrips. You'll need to change the paths to point to wherever the datasets are located.

For training NeXT-S/NeXT-B/NeXT-L of one scene on Blender dataset, run the following scrips:

cd jaxnerf
# training
sh scripts/train_next_blender.sh chair next_s   # next_s/next_b/next_l
# evaluation
sh scripts/eval_next_blender.sh chair next_s   # next_s/next_b/next_l

For training NeXT-S/NeXT-B/NeXT-L of one scene on Multi-scale dataset, run the following scrips:

cd mipnerf
# training
sh scripts/train_next_multiblender.sh chair next_s   # next_s/next_b/next_l
# evaluation
sh scripts/eval_next_multiblender.sh chair next_s   # next_s/next_b/next_l

Results

Blender

Multi-scale Blender

Acknowledgements

Thanks for the open-source JaxNeRF and Mip-NeRF.