Learning to Simplify: Fully Convolutional Networks for Rough Sketch Cleanup

SIGGRAPH 2016


Reviews

Information

  • Paper topic: Images
  • Software type: Code
  • Able to run a replicability test: True
  • Replicability score: 4
  • Software language: Python
  • License: SOFTWARE LICENSE AGREEMENT ACADEMIC OR NON-PROFIT ORGANIZATION NONCOMMERCIAL RESEARCH USE ONLY
  • Build mechanism: Not applicable (python, Matlab..)
  • Dependencies: pytorch0.4.1/torchvision0.2.0/pillow
  • Documentation score {0,1,2}: 1
  • Reviewer: Julie Digne <julie.digne@liris.cnrs.fr>
  • Time spent for the test (build->first run, timeout at 100min): 30min

Source code information

Comments

The provided code implements both a 2018 paper and the 2016 paper. To be sure to run the 2016 paper, on line 10 replace model_gan.t7 by model_mse.t7 to load the provided pre-trained model . Not all figures input are given (the figs.sh script is for the 2018 paper) but the test.png corresponds to one of the example in Fig. 01. My result was more blurry than the one in the paper, and so were the results with other test images I could run.

If you want to contribute with another review, please follow these instructions.

Please consider to cut/paste/edit the raw JSON data attached to this paper.