ALIGNet in Torch

[Project] [Arxiv]
This is our Torch implementation for ALIGNet, which is a network trained to align pairs of shapes without needing ground-truth warp fields for supervision. For example:

where the pink region is a missing piece in the target shape. The estimated alignments are oblivious to missing pieces in the partial shape.

The code was written by Rana Hanocka with support from Noa Fish and Zhenhua Wang.

This repo is still under active development

Setup

Prerequisites

• Linux (tested on Ubuntu 16.04, 14.04 and Linux Mint)
• NVIDIA GPU + CUDA (tested on cuda8 and cuda7.5) should also work on CPU

Getting Started

• Clone this repo:

git clone https://github.com/ranahanocka/ALIGNet.git
cd ALIGNet

• Install all dependencies: Torch-7, STN, HDF5:

sudo ls
chmod +x install.sh
./install.sh

Training

• download the data

chmod +x download_data.sh
./download_data.sh

• run training

th main.lua -data /path/to/data

Citation

If you use this code for your research, please cite our paper

@article{hanocka2018alignet,
  title={ALIGNet: Partial-Shape Agnostic Alignment via Unsupervised Learning},
  author={Hanocka, Rana and Fish, Noa and Wang, Zhenhua and Giryes, Raja and Fleishman, Shachar and Cohen-Or, Daniel},
  journal={arXiv preprint arXiv:1804.08497},
  year={2018}}

Acknowledgments

The code design and multithreading data loading capabilities used in this code were adopted from imagenet-multiGPU.torch.