Official code repository for the paper:

Generative Modelling of BRDF Textures from Flash Images [SIGGRAPH Asia, 2021]

Henzler, Deschaintre, J. Mitra, Ritschel

[Paper] [Project page]

The dataset is stored under flash_images and contains 306 train folders and 116 test folders (including images from Aitalla et al).

conda env create --name neuralmaterial --file=environment.yml

For training please run

python scripts/train.py

The default config is located at config/config_default.yaml.

Note, <test_id> is a relative path in the flash_images/test and <weight_name> is exptected to be located in the trainings/<run_name> folder.

In order to synthesise given flash images located in the test folder please run

python scripts/test.py --model <model_path> --test_image_id <test_id> --finetune <bool>

Given two images for interpolation please run

python scripts/interpolate.py --model <model_path> --weights1 <weight_name1> --weights2 <weight_name2> --test_image_id1 <test_id1> --test_image_id2 <test_id2>

If you would like to use fine-tuned weights please run the scripts/test.py command above in order to retrieve them.

Run the file run_examples.sh to synthesise / interpolate a few examples.

If you find our work useful in your research, please cite:

@article{henzler2021neuralmaterial,
  title={Generative Modelling of BRDF Textures from Flash Images},
  author={Henzler, Philipp and Deschaintre, Valentin and Mitra, Niloy J and Ritschel, Tobias},
  journal={ACM Trans Graph (Proc. SIGGRAPH Asia)},
  year={2021},
  volume={40},
  number={6},
}

If you have any questions, please email Philipp Henzler at philipp.henzler@cs.ucl.ac.uk.