Usage of https://github.com/davidsandberg/facenet/ pre-trained model for faces identification
- tensorflow
- opencv with python bindings (cv2)
- numpy
- sklearn for t-SNE and olivetti dataset
- matplotlib
- jupyter notebook for running .ipynb examples
- go to models folder and write the full paths to checkpoint file in the same folder
- put lfw dataset to data folder
- open lfw.ipynb or olivetti.ipynb for examples
- add confidence of prediction
- add flask server for incoming requests
- store embeddings in a file and load when start a server
- add data loader for lfw
- fix the problem due to which I need to reload the whole notebook to rerun cell with model loading