CFF : Convolutional Face Finder

C. Garcia, M. Delakis, University of Crete, 2004

We present a novel face detection approach based on a convolutional neural architecture, designed torobustly detect highly variable face patterns, rotated up to +/- 20 degrees in image plane and turned up to +/-60 degrees, in complex realworld images. The proposed system automatically synthesizes simple problem-specific feature extractors from a training set of faceand nonface patterns, without making any assumptions or using any hand-made design concerning the features to extract or the areasof the face pattern to analyze. The face detection procedure acts like a pipeline of simple convolution and subsampling modules thattreat the raw input image as a whole. We therefore show that an efficient face detection system does not require any costly localpreprocessing before classification of image areas. The proposed scheme provides very high detection rate with a particularly low levelof false positives, demonstrated on difficult test sets, without requiring the use of multiple networks for handling difficult cases. We present extensive experimental results illustrating the efficiency of the proposed approach on difficult test sets and including an in-depth sensitivity analysis with respect to the degrees of variability of the face patterns.