IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,
Neural Network-Based Face Detection
Henry A. Rowley,
Student Member, IEEE,
Shumeet Baluja, and Takeo Kanade,
—We present a neural network-based upright frontal face detection system. A retinally connected neural network examines
small windows of an image and decides whether each window contains a face. The system arbitrates between multiple networks to
improve performance over a single network. We present a straightforward procedure for aligning positive face examples for training.
To collect negative examples, we use a bootstrap algorithm, which adds false detections into the training set as training progresses.
This eliminates the difficult task of manually selecting nonface training examples, which must be chosen to span the entire space of
nonface images. Simple heuristics, such as using the fact that faces rarely overlap in images, can further improve the accuracy.
Comparisons with several other state-of-the-art face detection systems are presented, showing that our system has comparable
performance in terms of detection and false-positive rates.
—Face detection, pattern recognition, computer vision, artificial neural networks, machine learning.
N this paper, we present a neural network-based algo-
rithm to detect upright, frontal views of faces in gray-
The algorithm works by applying one or
more neural networks directly to portions of the input im-
age and arbitrating their results. Each network is trained to
output the presence or absence of a face. The algorithms
and training methods are designed to be general, with little
customization for faces.
Many face detection researchers have used the idea that
facial images can be characterized directly in terms of pixel
intensities. These images can be characterized by probabil-
istic models of the set of face images , ,  or im-
plicitly by neural networks or other mechanisms , ,
, , , , , . The parameters for these
models are adjusted either automatically from example
images (as in our work) or by hand. A few authors have
taken the approach of extracting features and applying ei-
ther manually or automatically generated rules for evalu-
ating these features , .
Training a neural network for the face detection task is
challenging because of the difficulty in characterizing proto-
typical “nonface” images. Unlike face
, in which
the classes to be discriminated are different faces, the two
classes to be discriminated in face
are “images con-
taining faces” and “images not containing faces.” It is easy to
1. An interactive demonstration of the system is available on the World
Wide Web at
, which allows anyone to
submit images for processing by the face detector, and to see the detection