zchen06/deep3dmm

Official repository for the CVPR 2021 paper "Learning Feature Aggregation for Deep 3D Morphable Models"

Deep3DMM

Official repository for the CVPR 2021 paper Learning Feature Aggregation for Deep 3D Morphable Models.

Requirements

This code is tested on Python 3.7 and Pytorch versoin 1.4 with CUDA 10.0. Requirments can be install by running

  pip install -r requirements.txt

Install mesh processing libraries from MPI-IS/mesh.

Train

To start the training, follow these steps

1. Download the registered data from the COMA and/or DFAUST.

2. Update default config file, default.cfg as needed, especially data_dir path.

3. Run the training of Deep3DMM by

   python main.py -m ComaAtt

   Note that the 'sliced' dataset split is used by default.

Evaluation

Run the evaluation by

 python main.py -m ComaAtt --eval

Note that the checkpoint with best validation accuracy is evaluated by default.

Acknowledgement

This implementation is built upon the Pytorch implementation of COMA (Link). We also build our Deep3DMM with spiral convolution based on the implementation of Neural3DMM. Many thanks to the authors for releasing the source code.

License

This code is free for non-commerical purposes only. For commercial usage, please contact the authors for more information.

Cite

Please consider citing our work if you find it useful:

Zhixiang Chen and Tae-Kyun Kim, "Learning Feature Aggregation for Deep 3D Morphable Models", CVPR, 2021.

deep3dmm's Issues

Is there a pre-training model to verify？thanks a lot.

Deep3DMM with spiral convolution？

hi，Could you give me some tips on how to build Deep3DMM with spiral convolution mentioned in the paper？Thanks

When I run the training code after download COMA dataset, it pop out below error message, would you please help me? thanks

RuntimeError: The size of tensor a (29990) must match the size of tensor b (5023) at non-singleton dimension 0

About layer in PYG2.0.2

Hi, I notice that the code can only run on PYG-1.4.3, do you have any plans to update it to PYG-2.0.2?