Skip to content

Latest commit

 

History

History
38 lines (30 loc) · 1.61 KB

dataset_preparation.md

File metadata and controls

38 lines (30 loc) · 1.61 KB

Data Preparation

We provide some tips for XRNerf data preparation in this file.

Getting Data

Dataset Organization

It is recommended to symlink the dataset root to $PROJECT/data. If your folder structure is different, you may need to change the corresponding paths in config files.

xrnerf
├── xrnerf
├── docs
├── configs
├── test
├── extensions
├── data
│   ├── nerf_llff_data
│   ├── nerf_synthetic
│   ├── multiscale
│   ├── ...

Dataset Download

  1. Download nerf_synthetic and nerf_llff_data from here, and put it under xrnerf/data
  2. Credit to NSVF authors for providing their datasets, read introductions here
  3. For mip-nerf training, you can generate the multiscale dataset used in the paper by running the following command, python tools/convert_blender_data.py --blenderdir /data/nerf_synthetic --outdir data/multiscale
  4. For the training of NeuralBody, please download the dataset from here.
  5. For the training of Animatable NeRF, please download the dataset from here.