Skip to content

Latest commit

 

History

History
30 lines (23 loc) · 1.59 KB

README.md

File metadata and controls

30 lines (23 loc) · 1.59 KB
Raw

Age Progression and Regression with Spatial Attention Modules

Sample age progression and regression results obtained via the proposed model

This repository contains the inference code for paper Age Progression and Regression with Spatial Attention Modules presented in AAAI2020.

Environment requirements

  • Python 2.7.5
  • Pytorch 0.3.1
  • CUDA 8.0

Usage

  1. Put your test images under the folder ./datasets/<name_of_dataset>. Then write a text file with each line containing <name_of_image> and <index_of_age_group>, seperated by a space character. Example input images and list files are provided.
  2. Please download pre-trained models (two generator networks) and place them under ./checkpoint.
  3. Modify test.sh as it fits your environment.
  4. run test.sh, and the generation results will be save into ./results. You could modify the corresponding code in age_cycle_gan_model.py to also generate attention maps.

Citation

@inproceedings{Li2020AgePA,
title={Age Progression and Regression with Spatial Attention Modules},
author={Qi Li and Yunfan Liu and Zhenan Sun},
booktitle={AAAI},
year={2020}
}

Acknowledgement

The framework of this project has borrowed from CycleGAN and GANimation. We appreaciate their great work!

Note

This repo is still under construction, please use with care.