Repository files navigation

PyTorch implementation ofLow-Resolution Face Recognition Based on Identity-Preserved Face Hallucination.

If you find this work useful for your research, please consider cite our paper:

@inproceedings{lai2019low,  title={Low-Resolution Face Recognition Based on Identity-Preserved Face Hallucination},  author={Lai, Shun-Cheung and He, Chen-Hang and Lam, Kin-Man},  journal={IEEE International Conference on Image Processing},  pages={1173-1177},  year={2019},  month={Sep.}}

• Python 3 (Anaconda installation is recommended)
• numpy
• PyTorch >= 0.4.1
• torchvision
• OpenCV 3
• tqdm (progress bar):pip install tqdm

Tested environment: Ubuntu 16.04 with Python 3.6.5, OpenCV 3.4.1, PyTorch 0.4.1 (CUDA 9.2 & cuDNN 7.1)

Clone this repository

git clone https://github.com/johnnysclai/SR_LRFRcd SR_LRFR/

To conduct low-resolution face verification, please download and extract the LFW database and 6,000 pairs file fromhere. Or, you just run the following commands:

mkdir datasetscd datasets/wget http://vis-www.cs.umass.edu/lfw/lfw.tgztar -xvzf lfw.tgzcd ../datawget http://vis-www.cs.umass.edu/lfw/pairs.txtcd ..

Now, you have the LFW database indatasets/lfw/ and the 6,000 pairs filedata/pairs.txt. We have usedMTCNN to detect five facial landmarks, which are saved indata/LFW.csv.

Extract the pre-trained checkpoints (models are compressed into parts as they are too large):

cd pretrained/# Extract edsr_baseline.pth and edsr_lambda0.5.pth7z x edsr_baseline.7z.0017z x edsr_lambda0.5.7z.001cd ..

Run the following commands to obtain the face verification results from our pre-trained models:

cd src/bash lfw_verification.sh

You should be able to get the following results:

*Note thatSphereFace-20 (SFace) model is converted from theofficial released model usingextract-caffe-params.

Please download the CelebA dataset fromhere and extract it to../../../Datasets/.

Train with single GPU:python train.py --gpu_ids=0

Train with two GPU:python train.py --gpu_ids=0,1

Final model will be saved in./checkpoint/master/

• LFW:project page andpaper
• MTCNN:project page,paper andcode
• SphereFace:paper andcode
• EDSR:paper andcode
• extract-caffe-params:https://github.com/nilboy/extract-caffe-params