Authors:
Andreas Savakis
and
David Higgs
Affiliation:
Rochester Institute of Technology, United States
Keyword(s):
Face detection, parts-based, multiple views, neural network, Bayesian network.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Data Manipulation
;
Health Engineering and Technology Applications
;
Human-Computer Interaction
;
Methodologies and Methods
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Pattern Recognition
;
Physiological Computing Systems
;
Sensor Networks
;
Soft Computing
Abstract:
This paper presents a parts-based approach to face detection, that is intuitive, easy to implement and can be used in conjunction with other image understanding operations that use prominent facial features. Artificial neural networks are trained as view-specific parts detectors for the eyes, mouth and nose. Once these salient facial features are identified, results for each view are integrated through a Bayesian network in order to reach the final decision. System performance is comparable to other state-of the art face detection methods while providing support for different view angles and robustness to partial occlusions.