Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/lyatdawn/LightCNN-mxnet

mxnet implementation of LightCNN in the paper "A Light CNN for Deep Face Representation with Noisy Labels" (https://arxiv.org/abs/1511.02683)
https://github.com/lyatdawn/LightCNN-mxnet

lightcnn mxnet

Last synced: about 2 months ago
JSON representation

mxnet implementation of LightCNN in the paper "A Light CNN for Deep Face Representation with Noisy Labels" (https://arxiv.org/abs/1511.02683)

Awesome Lists containing this project

README 

        
# Light CNN for Deep Face Recognition

* mxnet implementation of LightCNN in the paper "A Light CNN for Deep Face Representation with Noisy Labels" [[Paper]](https://arxiv.org/abs/1511.02683)

* Borrowed code and ideas from yxu0611's Tensorflow-implementation-of-LCNN: https://github.com/yxu0611/Tensorflow-implementation-of-LCNN.



## Install Required Packages

First ensure that you have installed the following required packages:

* mxnet0.12.0 ([instructions](http://mxnet.incubator.apache.org/install/index.html)). Maybe other version is ok.

* Opencv ([instructions](https://github.com/opencv/opencv)). Here is opencv-2.4.9.



See requirements.txt for details.



## Datasets

* In the implementation of the LightCNN, MsCelebV1-Faces-Aligned subset as the training data and validating data. Utilize the 10K, 70K [download](https://drive.google.com/open?id=1-jC6E2_gtuVLJxtITzSSX4ACKlOiGSat) to generate the subset of the MsCelebV1-Faces-Aligned. Then, use the mxnet tools im2rec.py to get the Record file, the image size is 144x144. When training LightCNN, all the training data are converted to gray-scale images and randomly croped to 128x128.



## Training a Model

* Run the following script to train and validate the model.

```shell

python train.py

```

You could change some arguments in the train.py, like num_epochs, gpus. If want to get the good results and save training time, there are some points should note:

 - For weights initialization,  use "xavier" initialization for "weight" of every layer, the "bias" will initializate to constant value "0.1". See **common/fit.py** initializer for details.

 - For learning rate,  use "AdamOptimizer" as optimization method, and initial learning rate is 0.00001. So in mxnet, you can set the initial learning rate is 0.0001, lr_step_epochs is '0, 100', in this way, the initial learning rate is 0.00001 in epoch 0.



 ## Extract features of LFW

 * Run the following script to extract every image's feature, moreover save the features and image path to MATLAB data.

 ```shell

 cd extract_features

 python extract_features.py

 ```

 The script will load the trained LightCNN model to generate fc layer output, w.t. the feature of a image. You could change the arguments in test.sh depend on your machine config. You could download the aligned LFW datasets and lfw_patch_part.txt from here [download](https://drive.google.com/open?id=1Xq6wITloANkIv66eysMw3ZwW1fv5-adM).



 ## Face Verification

 * Utilize MATLAB to run the lfw_eval.m to verificate a pair of images. You must generate the features of the aligned LFW datasets in advance. See **lfw_verification-matlab/lfw_eval.m** for details. 



## Downloading trained model

* Pretrained model: [download](https://drive.google.com/open?id=1-6g3tWrUfZdOofXgcQttUg1x5lhyuKBK). Training about 50 epoches use 10k people, the loss has converged. But the accuracy on the aligned LFW datasets is not high, about 85%. 



* Some improved solutions:

 1. Change the optimizer method, like: when loss doesn't decrease, manually decrease it to 10 time smaller.

 2. Improved by manaully aligning the images which are mis-algined in LFW datasets.

 3. Utilize a new feature of a image, for example use mfm_fc1_output as a image's feature, it will increase the accuracy in LFW.

 4. Improved by using metric learning method for similarity caculation, not just cos value.