Face Recognition using Convolutional Neural Networks (CNN)
Used a convolutional neural network for recognizing faces of five people: Tom Cruise, Donald Trump, Julia Roberts, Lady_A and Tom Hanks. Results prove that deep learning can be used for face recognition when implemented to account for it's limitations.
Training of the NIN convolutional network is very sensitive to quality of training dataset, preprocessing of training dataset and solver parameters.
Face recognition with NIN CNN with spatial pooling. The key advantage of spatial pooling is that it can generate a fixed-length representation regardless of the input image size/scale. Spatial pooling pools input feature map in local spatial bins with sizes proportional to the image size, resulting in a fixed number of bins. This is different from the sliding window pooling in the previous deep networks, where the number of sliding windows depends on the input size.
See more info at: https://www.linkedin.com/pulse/face-recognition-machine-learning-sarabjeet-singh https://www.linkedin.com/pulse/face-recognition-machine-learning-part-2-sarabjeet-singh