/FlattenAnything

Flatten Anything: Unsupervised Neural Surface Parameterization

Primary LanguageC++

Flatten Anything: Unsupervised Neural Surface Parameterization (NeurIPS-2024)

This is the official implementation of [Flatten Anything Model (FAM)], an unsupervised neural architecture to achieve global free-boundary surface parameterization via learning point-wise mappings between 3D points on the target geometric surface and adaptively-deformed UV coordinates within the 2D parameter domain. This code has been tested with Python 3.9, PyTorch 1.10.1, CUDA 11.1 and cuDNN 8.0.5 on Ubuntu 20.04.

Instruction

  • Within cdbs/CD/, run python setup.py install for compilation.

  • Within cdbs/EMD/, run python setup.py install and cp build/lib.linux-x86_64-cpython-39/emd_cuda.cpython-39-x86_64-linux-gnu.so . for compilation.

  • A demo script for running FAM is provided in a jupyter notebook scripts/main.ipynb.