Are you sure you want to delete this access key?
Legend |
---|
DVC Managed File |
Git Managed File |
Metric |
Stage File |
External File |
Legend |
---|
DVC Managed File |
Git Managed File |
Metric |
Stage File |
External File |
A PyTorch implementation of A Light CNN for Deep Face Representation with Noisy Labels from the paper by Xiang Wu, Ran He, Zhenan Sun and Tieniu Tan. The official and original Caffe code can be found here.
100% - EER
on LFW achieves 99.43%.100% - EER
on LFW achieves 99.40%.100% - EER
on LFW obtains 98.70%.Download face dataset such as CASIA-WebFace, VGG-Face and MS-Celeb-1M.
NOTE :- THE DATASET POINTED ABOVE HAS BEEN RETRACTED BY THE AUTHORS AND IS NO LONGER AVAILABLE.
All face images are converted to gray-scale images and normalized to 144x144 according to landmarks.
According to the five facial points, we not only rotate two eye points horizontally but also set the distance between the midpoint of eyes and the midpoint of mouth(ec_mc_y), and the y axis of midpoint of eyes(ec_y) .
The aligned LFW images are uploaded on Baidu Yun.
THE ABOVE BAIDU YUN LINK IS HARD TO ACCESS (WILL APPRECIATE CONTRIBUTIONS WITH IT)
Dataset | size | ec_mc_y | ec_y |
---|---|---|---|
Training set | 144x144 | 48 | 48 |
Testing set | 128x128 | 48 | 40 |
train.py
as a flag or manually change them.python train.py --root_path=/path/to/your/datasets/ \
--train_list=/path/to/your/train/list.txt \
--val_list=/path/to/your/val/list.txt \
--save_path=/path/to/your/save/path/ \
--model="LightCNN-9/LightCNN-29" --num_classes=n
load_imglist.py
. Or you can use torchvision.datasets.ImageFolder
to load your datasets.num_classes
denotes the number of identities in your training dataset.num_classes
is 80013.python extract_features.py --resume=/path/to/your/model \
--root_path=/path/to/your/datasets/ \
--img_list=/path/to/your/list.txt \
--save_path=/path/to/your/save/path/ \
--model="LightCNN-9/LightCNN-29/LightCNN-29v2"\
--num_classes=n (79077 for LightCNN-9/LightCNN-29, 80013 for LightCNN-29v2)
vlfeat
or sklearn
to evaluate the features on ROC and obtain EER
and TPR@FPR
for your testing datasets.train.py
.The Light CNN performance on lfw 6,000 pairs.
Model | 100% - EER | TPR@FAR=1% | TPR@FAR=0.1% | TPR@FAR=0 |
---|---|---|---|---|
LightCNN-9 | 98.70% | 98.47% | 95.13% | 89.53% |
LightCNN-29 | 99.40% | 99.43% | 98.67% | 95.70% |
LightCNN-29v2 | 99.43% | 99.53% | 99.30% | 96.77% |
LightCNN v4 | 99.67% | 99.67% | 99.57% | 99.27% |
The Light CNN performance on lfw BLUFR protocols
Model | VR@FAR=0.1% | DIR@FAR=1% |
---|---|---|
LightCNN-9 | 96.80% | 83.06% |
LightCNN-29 | 98.95% | 91.33% |
LightCNN-29v2 | 99.41% | 94.43% |
The Light CNN performance on MegaFace
Model | Rank-1 | TPR@FAR=1e-6 |
---|---|---|
LightCNN-9 | 65.782% | 76.288% |
LightCNN-29 | 72.704% | 85.891% |
LightCNN-29v2 | 76.021% | 89.740% |
If you use our models, please cite the following paper:
@article{wu2018light,
title={A light CNN for deep face representation with noisy labels},
author={Wu, Xiang and He, Ran and Sun, Zhenan and Tan, Tieniu},
journal={IEEE Transactions on Information Forensics and Security},
volume={13},
number={11},
pages={2884--2896},
year={2018},
publisher={IEEE}
}
Press p or to see the previous file or, n or to see the next file
Are you sure you want to delete this access key?
Are you sure you want to delete this access key?
Are you sure you want to delete this access key?
Are you sure you want to delete this access key?