Face Recognition with ColorFeret and LFW Datasets using different VGG architectures and Transfer Learning/Finetuning

This repository is very disorganized as you can probably guess from the existence of 14 different branches and of a very old project that I and one other student collaborator worked on 2 years back without both of us having proper knowledge of Git . You can look through all the branches but I admit that it's super messy now with alls sorts of protobuf and prototxt files that we used for our project back then. It also contains several bash scripts for preprocessing, training, validation and testing, and also several python scripts for interfacing i.e. doing training and validation. We were absolutely raw and new into both Deep Learning and Caffe back then.