-
Notifications
You must be signed in to change notification settings - Fork 14
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
1 changed file
with
31 additions
and
1 deletion.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,2 +1,32 @@ | ||
# mxnet-E2FAR | ||
MXNET/Gluon Implementation of End-to-end 3D Face Reconstruction with Deep Neural Networks | ||
This is a MXNet/Gluon Implementation of End-to-end 3D Face Reconstruction with Deep Neural Networks. | ||
|
||
1. Download VGG-Face and convert it to the mxnet-weights by running the caffe_converter: | ||
``` | ||
python $MXNET/tools/caffe_converter/convert_model.py prototxt weights params_name | ||
``` | ||
Put the weights into the folder ```ckpt/VGG-Face``` | ||
|
||
2. Prepare the dataset | ||
|
||
3. For train your dataset, you may need to change the ```dataset``` in the main code to fit your dataset | ||
|
||
Run the code: | ||
``` | ||
# fine-tune the branch and fully connected layers | ||
python E2FAR.py --pretrained --freeze --epoch 10 | ||
# fine-tune whole network | ||
python E2FAR.py --start_epoch 10 | ||
``` | ||
|
||
If you use this code, pls mention this repo and cite the paper: | ||
``` | ||
@InProceedings{Dou_2017_CVPR, | ||
author = {Dou, Pengfei and Shah, Shishir K. and Kakadiaris, Ioannis A.}, | ||
title = {End-To-End 3D Face Reconstruction With Deep Neural Networks}, | ||
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)}, | ||
month = {July}, | ||
year = {2017} | ||
} | ||
``` |