/cleanir

Official implementation of CLEANIR

Primary LanguageJupyter NotebookGNU General Public License v3.0GPL-3.0

CLEANIR: Controllable Attribute-Preserving Natural Identity Remover

This is the official implementation of CLEANIR: Controllable Attribute-Preserving Natural Identity Remover

CLEANIR Demo

Requirements

  • face-recognition>=1.2.3
  • opencv-python>=4.1.0.25
  • Keras>=2.2.4
  • tensorflow-gpu>=1.13.0rc0+nv
  • matplotlib>=3.0.2
  • numpy>=1.14.5
  • pandas>=0.23.0
  • tqdm>=4.32.2
  • gdown>=3.10.2
  • keras-facenet>=0.1a5 (if you want to run codes for evaluation on de-identification)
  • azure-cognitiveservices-vision-face>=0.4.0 (if you want to run codes for evaluation on preserving facial emotion)

Testing environment

  • Ubuntu 16.04
  • Python 3.5.2
  • Keras 2.2.4 (backend=TensorFlow 1.13.0-rc0)

Usage

Please check CLEANIR_notebook.ipynb