TL;DR

Image rendering that can even run on your phone with a batch run on VESSL.

Description

Neural Radiance Fields (NeRFs) have impressive image synthesis capabilities for 3D scenes. This paper introduces a new NeRF representation using textured polygons that can efficiently synthesize images using standard rendering pipelines. By incorporating a z-buffer, which assigns features to each pixel, and utilizing a view-dependent MLP in a fragment shader, the final pixel colors are produced. This approach enables NeRFs to be rendered with the traditional polygon rasterization pipeline, achieving interactive frame rates on various compute platforms.

YAML

name: mobilenerf
description: "Image rendering that can even run on your phone with a batch run on VESSL."
resources:
  cluster: aws-apne2
  preset: v1.v100-1.mem-52
image: quay.io/vessl-ai/ngc-pytorch-kernel:22.12-py3-202301160809
run:
  - workdir: /root
    command: |
       unzip /root/datasets/nerf_synthetic.zip -d /datasets/
       git clone <https://github.com/treasuraid/jax3d.git>
  - workdir: /root/jax3d/jax3d/projects/mobilenerf
    command: |
       pip3 install -r requirements.txt 
       pip install jaxlib==0.1.69+cuda111 -f <https://storage.googleapis.com/jax-releases/jax_cuda_releases.html>
       python stage1.py
import:
  /root/datasets/: s3://vessl-public-apne2/vessl_run_datasets/cvpr_candidates/nerf_synthetic.zip

Images

Input

Output