Rotation Invariant Neural Network-Based Face Detection - Robotics Institute Carnegie Mellon University

Rotation Invariant Neural Network-Based Face Detection

Henry Rowley, Shumeet Baluja, and Takeo Kanade
Tech. Report, CMU-CS-97-201, Computer Science Department, Carnegie Mellon University, December, 1997

Abstract

In this paper, we present a neural network-based face detection system. Unlike similar systems which are limited to detecting upright, frontal faces, this system detects faces at any degree of rotation in the image plane. The system employs multiple networks; the first is a "router" network which processes each input window to determine its orientation and then uses this information to prepare the window for one or more "detector" networks. We present the training methods for both types of networks. We also perform sensitivity analysis on the networks, and present empirical results on a large test set. Finally, we present preliminary results for detecting faces which are rotated out of the image plane, such as profiles and semi-profiles.

BibTeX

@techreport{Rowley-1997-14549,
author = {Henry Rowley and Shumeet Baluja and Takeo Kanade},
title = {Rotation Invariant Neural Network-Based Face Detection},
year = {1997},
month = {December},
institute = {Carnegie Mellon University},
address = {Pittsburgh, PA},
number = {CMU-CS-97-201},
}