Learning Deep Face Representation

摘要

Face representation is a crucial step of face recognition systems. An optimal face representation should be discriminative, robust, compact, and very easy to implement. While numerous hand-crafted and learning-based representations have been proposed, considerable room for improvement is still present. A very easy-to-implement deep learning framework for face representation is presented. The framework bases on pyramid convolutional neural network (CNN). The pyramid CNN adopts a greedy-filter-and-down-sample operation, which enables the training procedure to be very fast and computation efficient. In addition, the structure of Pyramid CNN can naturally incorporate feature sharing across multi-scale face representations, increasing the discriminative ability of resulting representation.

主权项

1. A computer-implemented method for training a pyramid convolutional neural network (CNN) comprising at least N shared layers where N≧2 and at least one unshared network coupled to the Nth shared layer, the method comprising:
training CNN levels 1 to N in that order, wherein CNN level n comprises an input for receiving face images, the first n shared layers of the pyramid CNN, the unshared network of the pyramid CNN, and an output producing representations of the face images; wherein the input is coupled to a first of the n shared layers; each shared layer includes convolution, non-linearity and down-sampling; an nth of the n shared layers is coupled to the unshared network; and the unshared network is coupled to the output; wherein training CNN level n comprises:
presenting face images to the input, each face image producing a corresponding representation at the output;processing the representations to produce estimates of a metric, for which actual values of the metric are known; andadapting the nth shared layer and the unshared network based on the estimates of the metric compared to the actual values of the metric.