Bernhard Fröba and Christian Küblbeck,

"Orientation Template Matching for Face Localization in Complex Visual Scenes",

International Conference on Image Processing ICIP2000, 251-254, 2000

An important topic in face recognition as well as in video coding or multi-modal human machine interfaces is the automatic localization of faces or head-and-shoulder regions in visual scenes. The algorithms therefore should be computationally efficient and robust against distortions like varying lighting conditions. This paper describes a novel method for segmenting frontal head and shoulder views of persons from grey level images. The segmentation is done by oriented template correlation. This matching method only depends on edge information, especially the orientation of the edges. We describe the generation of a statistical model of edge orientation within a human face from a training sample. In the matching stage we calculate the probability for a face at the current image position using this model. The detection capabilities of the presented algorithm are evaluated on a large database of 1114 images each containing one or more faces.

Bernhard Fröba and Christian Küblbeck, 

"Face Detection and Tracking using Edge Orientation Information",

SPIE Visual Communications and Image Processing, 583-594, 2001

An important topic in face recognition as well as in video coding or multi-modal human machine interfaces is the automatic detection of faces or head-and-shoulder regions in visual scenes. The algorithms therefore should be computationally fast enough to allow an online detection and parallel processing of the detected objects. In this paper we describe our ongoing work on face detection using an approach that models the face appearance by edge orientation information. We will show that edge orientation is a powerful local image feature to describe objects like faces or body parts. We will present a simple and efficient method for template matching and object modeling based solely on edge orientation information. We also show how to obtain an optimal face model in the edge orientation domain from a set of training images. Unlike many approaches that model the grey level appearance of the face our approach is computationally very fast. It takes less than 0.1 seconds on a Pentium II 500MHz for a 190x140 image to be processed using a multi-resolution search with three resolution levels. We demonstrate the capability of our detection method on an image database of 7000 images taken from more than 200 different people. The variations in head size, lighting and background are considerable. The obtained detection rate is more than 97% on that database. We have also extended the algorithm for face tracking. The tracking capabilities are shown using results from a real - time implementation of the proposed algorithm.

Bernhard Fröba and Walter Zink, 

"On the Combination of Different Template Matching Strategies for Fast Face Detection",

Multiple Classifier Systems (MCS'2001), 418-428, 2001

Computer-based face perception is becoming increasingly important for many applications like biometric face recognition, video coding or multi-model human-machine interaction. Fast and robust detection and segmentation of a face in an unconstrained visual scene is a basic requirement for all kinds of face perception. This paper deals with the integration of three simple visual cues for the task of face detction in grey level images. It is achieved by a combination of edge orientation matching, hough transform and an appearance based detection method. The proposed system is computationally efficient and has proved to be robust under a wide range of acquisition conditions like varying lighting, pixel noise and other image distortions. The detection capabilities of the presented algorithm are evaluated on a large database of 13122 images including the frontal-face set of the m2vts database. We achieve a detrection rate of over 91% on this databese while having only few false detects at the same time.

Bernhard Fröba and Christian Küblbeck,

"Real-Time Face Detection using Edge-Orientation Matching",

Audio- and Video-based Biometric Person Authentication (AVBPA'2001), 78-83, 2001

In this paper we describe our ongoing work on real-time face detection in grey level images using edge orientation information. We will show that edge orientation is a powerful local image feature to model objects like faces for detection purposes. We will present a simple and efficient method for template matching and object modeling based solely on edge orientation information. We also show how to obtain an optimal face model in the edge orientation domain from a set of training images. Unlike many approaches that model the grey level appearance of the face our approach is computationally very fast. It takes less than 0.08 seconds on a Pentium II 500MHz for a 320x240 image to be processed using a multi-resolution search with six resolution levels. We demonstrate the capability of our detection method on an image database of 17000 images taken from more than 2900 different people. The variations in head size, lighting and background are considerable. The obtained detection rate is more than 93% on that database.

Christian Küblbeck and Bernhard Fröba,

"Gesichtsdetektion mithilfe von Farb- und Formmodellen",

GfaI 6. Workshop Farbbildverarbeitung, 75-82, 2000, ISBN 3-9807029-4-4

In diesem Artikel wird ein Ansatz beschrieben, der auf farbbasierten Bildern Gesichter lokalisiert. Hierfür wird zunächst anhand einer Lernstichprobe ein Hautfarbenmodell unter Verwendung von Histogrammen oder von Gauß-Mixturen berechnet. Dieses wird dann dazu verwendet, in Testbildern hautfarbene Pixel zu lokalisieren. Anschließend werden zusammengehörige Pixel mithilfe von Wachstumsverfahren zu Regionen zusammengefasst (segmentiert). Schließlich wird anhand der Form der Regionen eine Klassifikation durchgeführt. Dazu werden parametrische Formmodelle, wie z.B. Ellipsen, wie auch Kantenmodelle von Gesichtern verwendet. In den Ergebnissen werden die verschiedenen Ansätze mit einander verglichen.

Christian Küblbeck, 

"Biometrische Analyse der Lippenbewegungen zur Unterstützung der akustischen Spracherkennung",

in "Biometrische Verfahren", Fachverlag Deutscher Wirtschaftsdienst, Veronika Nolde und Lothar Leger (Hrsg.), 348-353, 2002, ISBN 3-87156-464-8

Der Einsatz biometrischer Verfahren bietet sich nicht nur im Zusammenhang mit Sicherheitssystemen an, sondern kann auch im Bereich der Mensch-Technik-Interaktion eine tragende Rolle spielen. Der Beitrag beschreibt ausblickartig die Möglichkeiten, die sich durch die Analyse von Lippenbewegungen ergeben. Es werden Möglichkeiten der Merkmalserzeugung dargestellt. Vor allem in Verbindung mit der akustischen Erkennung ergeben sich Potenziale, da so die Beeinträchtigung durch Nebengeräusche ausgeräumt werden könnte.