An Android as frontend and Python as backend face recognition app used on attendance management
#Python #Flask #Tensorflow #Android #Tenserflow-lite
- Create a python virtual environment -> python3 -m virtualenv env
- Active env -> source env/bin/activate
- Install all required library -> python3 -m pip install -r requirements.txt (By mistake or any how sklearn not install by above command , use this - python3 -m pip install -U scikit-learn scipy matplotlib)
- First Record face of the person using python app - -> python3 face_generate.py -> It will ask you to enter person name, enter and when camera open let read your face. -> for better result, move your face on different direction to record
- After Record done, train dataset. It will regenerate new model with all recorded face. Face images will be on dataset named folder. -> python3 face_train.py -> It will create a new model file named "trained_knn_model.clf"
- Now keep run the API -> python3 face_api.py -> You can modify this file as per your requirements
- Change API URL on WebServices -> ConstantString.java page
- Might be you will need to manage camera rotation if you cant see bounding box in face... -> Check CameraActivity.java - line 276 -> You may need to modify as per your requirements. -> If face detctor will not false after detecting in DetectorActivity.java line 284 you will get result by late -> also because it is asynchronous and its keep calling.. -> If result is [] , mean face not matched -> You need to manage flag for this.
https://drive.google.com/file/d/1-z2bFYZ4LQO0cz5Pi9ULGNzOg2ppEKrL/view?usp=sharing