Facial Recognition Enhancement Using Deep Learning Techniques
Our aim in this paper is to increase the accuracy of existing facial recognition system on a comparative smaller dataset as per the requirements of present day. Namely in sensitive regions. The methodology that has been adopted is by combining more than one algorithm. Data is an important factor for learning a deep face representation, several research groups have been collecting datasets with images ranging from 90,000 to 2,600,000 labeled images. The feature detection capability of harr cascade along with Ada boost to fetch to Bilinear CNN so that on a comparative smaller dataset can produce comparative result as on bigger dataset.