Face-Recognition-master (1).zip

# Face Recognition using ARCFACE-Pytorch ## Introduction This repo contains face_verify.py and app.py which is able to perform the following task - - Detect faces from an image, video or in webcam and perform face recogntion. - app.py was used to deploy the project. ## Required Files - requirements.txt - pretrained model [IR-SE50 @ Onedrive](https://onedrive.live.com/?authkey=%21AOw5TZL8cWlj10I&cid=CEC0E1F8F0542A13&id=CEC0E1F8F0542A13%21835&parId=root&action=locate) or [Mobilefacenet @ OneDrive](https://onedrive.live.com/?authkey=%21AIweh1IfiuF9vm4&cid=CEC0E1F8F0542A13&id=CEC0E1F8F0542A13%21836&parId=root&o=OneUp). - Custom dataset - Newly Trained model (facebank.pth and names.npy) ## User Instruction After downloading the project first you have to install the following libraries. ### Installation You can install all the dependencies at once by running the following command from your terminal. ``` python $ pip install -r requirements.txt ``` ##### For the installation of torch using "pip" run the following command ``` python $ pip3 install torch===1.2.0 torchvision===0.4.0 -f https://download.pytorch.org/whl/torch_stable.html ``` ### Project Setup #### pre-trained model Although i provided the pretrained model in the <b> work_space/model </b> and <b> work_space/save </b> folder, if you want to download the models you can follow the following url: - [IR-SE50 @ BaiduNetdisk](https://pan.baidu.com/s/12BUjjwy1uUTEF9HCx5qvoQ) - [IR-SE50 @ Onedrive](https://onedrive.live.com/?authkey=%21AOw5TZL8cWlj10I&cid=CEC0E1F8F0542A13&id=CEC0E1F8F0542A13%21835&parId=root&action=locate) - [Mobilefacenet @ BaiduNetDisk](https://pan.baidu.com/s/1hqNNkcAjQOSxUjofboN6qg) - [Mobilefacenet @ OneDrive](https://onedrive.live.com/?authkey=%21AIweh1IfiuF9vm4&cid=CEC0E1F8F0542A13&id=CEC0E1F8F0542A13%21836&parId=root&o=OneUp) I have used the <b>IR-SE50</b> as the pretrained model to train with my custom dataset. You need to copy the pretrained model and save it under the <b> work_space/save </b> folder as <b> model_final.pth</b> #### Newly trained model In the <b> data/facebank </b> you will find a trained model named <b> "facebank.pth" </b> which contains the related weights and "names.npy" contains the corresponding labels of the users that are avialable in the facebank folder. For instance in this case the <b> facebank </b> folder will look like this :- facebank/ ---> Chandler ---> Joey ---> Monica ---> Phoebe ---> Pias ---> Rachel ---> Raihan ---> Ross ---> Samiur ---> Shakil ---> facebank.pth ---> names.npy If you have the "facebank.pth" and "names.npy" files in the <b>data/facebank</b> you can execute the following command to see the demo. ```python $ python app.py ``` and go to the following url from your web browser. ```url http://localhost:5000 ``` <hr> Note: If you want to run the inference on a video, download a video of related persons (Person that you trained the model with) and replace 0 in the line number 43 of <b> face_verify.py </b> with the path of your video file. For this code you can run the inference on any video of <b> Friends</b> tv series. <hr> #### Now if you want to train with your custom dataset, you need to follow the following steps. #### Dataset preparation First organize your images within the following manner- data/ raw/ name1/ photo1.jpg photo2.jpg ... name2/ photo1.jpg photo2.jpg ... ..... now run the following command ```python $ python .\create-dataset\align_dataset_mtcnn.py data/raw/ data/processed --image_size 112 ``` You will see a new folder inside the data directory named <b> "processed" </b> which will hold all the images that contains only faces of each user. If more than 1 image appears in any folder for a person, average embedding will be calculated. After executing the script new images for each user in the processed folder will look something like this. <p align="center"> <b> Cropped Images of faces </b> <img src ="http://muizzer07.pythonanywhere.com/media/files/Picture1.png"> </p> Copy all the folders of the users under the <b>data/processed</b> folder and paste in the <b>data/facebank</b> folder. Now to train with your dataset, you need to set <b> args.update == True </b> in line 35 of face_verify.py . After training you will get a new facebank.pth and names.npy in your data/facebank folder which will now only holds the weights and labels of your newly trained dataset. Once the training is done you need to reset <b> args.update==False</b>. However, if this doesn't work change the code in following manner- #### Old Code ```python if args.update: targets, names = prepare_facebank(conf, learner.model, mtcnn, tta = args.tta) print('facebank updated') else: targets, names = load_facebank(conf) print('facebank loaded') ``` #### New Code Only keep the follwing lines for training, once the training is done just replace it with the old code. ```python targets, names = prepare_facebank(conf, learner.model, mtcnn, tta = args.tta) print('facebank updated') ```` Or you can simply pass a command line arguement such as below if there is new data to train. ```python $python face_verify.py -u ``` Here the -u parse the command to update the facebank.pth and names.npy. Now you are ready to test the systen with your newly trained users by running- ```python $ python app.py ``` ### Note: You can train with custom dataset as many time as you want, you will only require any of the pre-trained model to train with your custom dataset to generate the <b>facebank.pth</b> and <b>names.npy</b>. Once you get this two files you are ready to test the face recogniton. <hr> ### Retrain the pre-trained model If you want to build a new pre-trained model like [IR-SE50 @ Onedrive](https://onedrive.live.com/?authkey=%21AOw5TZL8cWlj10I&cid=CEC0E1F8F0542A13&id=CEC0E1F8F0542A13%21835&parId=root&action=locate) and reproduce the result, you will need the large files which contains several dataset of faces under the <b> data/faces_emore </b>. To get these files, first you need to download the [MS1M](https://arxiv.org/abs/1607.08221) dataset from any of the following url- - [emore dataset @ Dropbox](https://www.dropbox.com/s/wpx6tqjf0y5mf6r/faces_ms1m-refine-v2_112x112.zip?dl=0) - [emore dataset @ BaiduDrive](https://pan.baidu.com/s/1eXohwNBHbbKXh5KHyItVhQ) After unzipping the downloaded file execute the following command. It will take few hours depending on your system configuration. ```python $ python prepare_data.py ``` After that you will see the following files to the "data/faces_emore" folder. faces_emore/ ---> agedb_30 ---> calfw ---> cfp_ff ---> cfp_fp ---> cfp_fp ---> cplfw ---> imgs ---> lfw ---> vgg2_fp To know the further training procedure you can see the details in this [InsightFace_Pytorch](https://github.com/TreB1eN/InsightFace_Pytorch) repository. ## References - [Arcface](https://arxiv.org/pdf/1801.07698.pdf) - [InsightFace_Pytorch](https://github.com/TreB1eN/InsightFace_Pytorch) - [The one with Face Recognition.](https://towardsdatascience.com/s01e01-3eb397d458d)