Machine Learning

NeRF Meta-Learning with PyTorch

nerf-meta is a PyTorch re-implementation of NeRF experiments from the paper “Learned Initializations for Optimizing Coordinate-Based Neural Representations”. Simply by initializing NeRF with meta-learned weights, we can achieve:

Environment

  • Python 3.8
  • PyTorch 1.8
  • NumPy, imageio, imageio-ffmpeg

Photo Tourism

Starting from a meta-initialized NeRF, we can interpolate between camera pose, focal length, aspect ratio and scene appearance. The videos below are generated with a 5 layer only NeRF, trained for ~100k iterations.

Data

Train and Evaluate

  1. Train NeRF on a single landmark scene using Reptile meta-learning:

    python tourism_train.py --config ./configs/tourism/$landmark.json

  2.  

     

     

    To finish reading, please visit source site