Update README.md

Author: xuexingyu24

README.md

@@ -1,30 +1,35 @@
 # MobileFaceNet_Tutorial_Pytorch
-This repo illustrates how to implement MobileFaceNet and Arcface for face recognition task.  
 
-## Demo
+* This repo illustrates how to implement MobileFaceNet and Arcface for face recognition task.
+* Pretrained model is posted for tests over picture, video and cam
+* Help document on how to implement MTCNN+MobileFaceNet is available
+* Scripts on transforming MXNET data records in [Insightface](https://github.com/deepinsight/insightface/wiki/Dataset-Zoo) to images are provided 
+* Scripts on train and evaluation of MobileFaceNet model are provided 
+
+## MobileFaceNet Video Demo
 
 <img src="images/ipy_pic/output.gif"  width="700" style="float: left;">
 
-## Test
+## Test over Picture, Video and Cam
 1. Test Picture
   ```
   python MTCNN_MobileFaceNet.py -img {image_path}
   ```
 2. Take Picture for Face Database
-* From cam
+* over cam
   ```
   python take_picture.py -n {name}
   ```   
-* From photo
+* over photo
   ```
   python take_ID.py -i {image_path} -n {name}
   ```
 3. Test Video
-* From cam
+* over cam
   ```
   python cam_demo.py
   ```
-* From video file
+* over video file
   ```
   python video_demo.py
   ```
@@ -33,8 +38,7 @@ This repo illustrates how to implement MobileFaceNet and Arcface for face recogn
   MobileFaceNet_Step_by_Step.ipynb
   ```
 ## Train
-Download training and evaluation data from [Model Zoo](https://github.com/deepinsight/insightface/wiki/Dataset-Zoo).
-Put images and annotation files into "data_set" folder. The structure should be arranged as follows:
+Download training and evaluation data from [Model Zoo](https://github.com/deepinsight/insightface/wiki/Dataset-Zoo). All training data has been cropped, aligned and resized as 112 x 112. Put images and annotation files into "data_set" folder. The structure should be arranged as follows:
   ```
   data_set/
               ---> AgeDB-30
@@ -43,11 +47,11 @@ Put images and annotation files into "data_set" folder. The structure should be
               ---> faces_emore_images
               ---> LFW
   ```
-1. Convert .bin and .rec file to images 
+1. The following script is provided to convert .bin and .rec file to images:
   ```
   python data_set/load_images_from_bin.py
   ```
-2. Generate annotation file 
+2. Generate the corresponding annotation files
   ```
   python data_set/anno_generation.py
   ```
@@ -59,7 +63,8 @@ Put images and annotation files into "data_set" folder. The structure should be
   ```
   MobileFaceNet_Training_Step_by_Step.ipynb
   ```
-The training results are shown below:
+The training results over faces_emore data (5822653 images / 85742 ids) are shown below:
+
 <table><tr>
 <td> <img src="images/ipy_pic/loss_train.png"  width="500" style="float: left;"> </td>
 <td> <img src="images/ipy_pic/accuracy_train.png"  width="500" > </td>
@@ -69,7 +74,7 @@ The training results are shown below:
   ```
   python Evaluation.py
   ``` 
-Here is the evaluation result. 'Flip' the image can be applied to encode the embedding feature vector with ~ 0.07% higer accuracy. L2 distance score slightly outperforms cos similarity (not necessarily the same trend for other cases, but it is what we conclude in this work) 
+Here is the evaluation result. 'Flip' the image could be applied to encode the embedding feature vector with ~ 0.07% higer accuracy. L2 distance score slightly outperforms cos similarity (not necessarily the same trend for other cases, but it is what we conclude in this work) 
 
 |  Eval Type     |   Score   |   LFW   | AgeDB-30 | CFP-FP 
 |:--------------:|:---------:|:-------:|:--------:|:-------
@@ -79,3 +84,8 @@ Here is the evaluation result. 'Flip' the image can be applied to encode the emb
 |UnFlip          |  Cos      |  99.45  |   95.65  |  93.10 
 
 Don't forget to star the repo if it is helpful for your research 
+
+## Reference 
+* https://github.com/deepinsight/insightface
+* https://github.com/wujiyang/Face_Pytorch
+* https://github.com/TreB1eN/InsightFace_Pytorch