Rotation invariant face detection using convolutional neural networks

Authors

Fok Hing Chi Tivive, University of Wollongong
Abdesselam Bouzerdoum, University of WollongongFollow

RIS ID

14186

Publication Details

F. Tivive & A. Bouzerdoum, "Rotation invariant face detection using convolutional neural networks", in International Conference on Neural Information Processing, 3-6 October, Hong Kong, China, Lecture Notes in Computer Science Series Neural Information Processing, vol. 4233, pp.260-269, 2006. L. Chan, I. King, J. Wang & D. Wang.

Abstract

This article addresses the problem of rotation invariant face detection using convolutional neural networks. Recently, we developed a new class of convolutional neural networks for visual pattern recognition. These networks have a simple network architecture and use shunting inhibitory neurons as the basic computing elements for feature extraction. Three networks with different connection schemes have been developed for in-plane rotation invariant face detection: fully-connected, toeplitz-connected, and binary-connected networks. The three networks are trained using a variant of Levenberg-Marquardt algorithm and tested on a set of 40,000 rotated face patterns. As a face/non-face classifier, these networks achieve 97.3% classification accuracy for a rotation angle in the range ±90⁰ and 95.9% for full in-plane rotation. The proposed networks have fewer free parameters and better generalization ability than the feedforward neural networks, and outperform the conventional convolutional neural networks.