This is the TensorFlow implementation of "Learning Adaptive Hierarchical Cuboid Abstractions of 3D Shape Collections" by Chun-Yu Sun, Qian-Fang Zou, Xin Tong, Yang Liu, SIGGRAPH Asia 2019. The code is released under the MIT license.

Pre-prequisites

    Python == 3.6
    TensorFlow == 1.12

To start the initial training, run

    $ python initial_training.py --log_dir /path/to/save/weights --cache_folder /path/to/save/tmp_results 

During training, the network outputs intermediate data into these folders:

• log_dir: Training logs which can be visualized using Tensorboard and network snapshots which can be used in evaluation.
• cache_folder: Cuboids visualization during training.

To test a trained model, run

    $ python initial_training.py --ckpt /path/to/snapshots --cache_folder /path/to/save/test_results --test

If you use our code for research, please cite our paper:

@article{sun2019abstraction,
  title     = {Learning Adaptive Hierarchical Cuboid Abstractions of 3D Shape Collections},
  author    = {Sun, Chunyu and Zou, Qianfang and Tong, Xin and Liu, Yang},
  journal   = {ACM Transactions on Graphics (SIGGRAPH Asia)},
  volume    = {38},
  number    = {6},
  year      = {2019},
  publisher = {ACM}
}

Please contact us (Chunyu Sun sunchyqd@gmail.com, Yang Liu yangliu@microsoft.com) if you have any problem about our implementation or request to all the datasets.