This package contains the working code for the method described in the paper "Sketch-based 3d shape retrieval using convolutional neural networks". 
Please notice that the code is developed for research projects. Please contact the author through fang.wang@nicta.com.au if you encountered any problem. 


=== Usage ===

Train the model using

  $ python shape_retrieval.py  train config/exp-shrec.cfg  

then, generate the features using

  $ python shape_retrieval.py  test config/exp-shrec.cfg

After the above step, the model and feature files should be created under [./cache/]. 
Then you are ready to run the matlab evaluation code to get the final scores (see demo.m as an example).


=== Citation ===

Please cite our work if you use this code in your research:

    @inproceedings{wang2015sketch,
      title={Sketch-based 3d shape retrieval using convolutional neural networks},
      author={Wang, Fang and Kang, Le and Li, Yi},
      booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
      pages={1875--1883},
      year={2015}
    }

Authors: 
    Fang Wang, NICTA and Australian National University
    Le Kang, University of Maryland
    Yi Li, NICTA, Australian National University, and Toyota Research Institute