Showing papers on "Face detection published in 1991"

Eigenfaces for recognition

Abstract: We have developed a near-real-time computer system that can locate and track a subject's head, and then recognize the person by comparing characteristics of the face to those of known individuals. The computational approach taken in this system is motivated by both physiology and information theory, as well as by the practical requirements of near-real-time performance and accuracy. Our approach treats the face recognition problem as an intrinsically two-dimensional (2-D) recognition problem rather than requiring recovery of three-dimensional geometry, taking advantage of the fact that faces are normally upright and thus may be described by a small set of 2-D characteristic views. The system functions by projecting face images onto a feature space that spans the significant variations among known face images. The significant features are known as "eigenfaces," because they are the eigenvectors (principal components) of the set of faces; they do not necessarily correspond to features such as eyes, ears, and noses. The projection operation characterizes an individual face by a weighted sum of the eigenface features, and so to recognize a particular face it is necessary only to compare these weights to those of known individuals. Some particular advantages of our approach are that it provides for the ability to learn and later recognize new faces in an unsupervised manner, and that it is easy to implement using a neural network architecture.

Face recognition using eigenfaces

Abstract: An approach to the detection and identification of human faces is presented, and a working, near-real-time face recognition system which tracks a subject's head and then recognizes the person by comparing characteristics of the face to those of known individuals is described. This approach treats face recognition as a two-dimensional recognition problem, taking advantage of the fact that faces are normally upright and thus may be described by a small set of 2-D characteristic views. Face images are projected onto a feature space ('face space') that best encodes the variation among known face images. The face space is defined by the 'eigenfaces', which are the eigenvectors of the set of faces; they do not necessarily correspond to isolated features such as eyes, ears, and noses. The framework provides the ability to learn to recognize new faces in an unsupervised manner. >

PICTION: a system that uses captions to label human faces in newspaper photographs

Abstract: It is often the case that linguistic and pictorial information are jointly provided to communicate information. In situations where the text describes salient aspects of the picture, it is possible to use the text to direct the interpretation (i.e., labelling objects) in the accompanying picture. This paper focuses on the implementation of a multi-stage system PICTION that uses captions to identify humans in an accompanying photograph. This provides a computationally less expensive alternative to traditional methods of face recognition. It does not require a pre-stored database of face models for all people to be identified. A key component of the system is the utilisation of spatial constraints (derived from the caption)in order to reduce the number of possible labels that could be associated with face candidates (generated by a face locator). A rule-based system is used to further reduce this number and arrive at a unique labelling. The rules employ spatial heuristics as well as distinguishing characteristics of faces (e.g., male versus female). The system is noteworthy since a broad range of AI techniques are brought to bear (ranging from natural-language parsing to constraint satisfaction and computer vision).

Interactive-time vision: face recognition as a visual behavior

Abstract: This thesis describes a vision system which performs face recognition as a specialpurpose visual task, or "visual behavior". In addition to performing experiments using stored face images digitized under a range of imaging conditions, I have implemented face recognition in a near-real-time (or "interactive-time") computer system which locates and tracks a subject's head and then recognize the person by comparing characteristics of the face to those of known individuals. The computational approach of this system is motivated by both biology and information theory, as well as by the practical requirements of interactive-time performance and accuracy. The face recognition problem is treated as an intrinsically two-dimensional recognition problem, taking advantage of the fact that faces are normally upright and thus may be described by a small set of 2-D characteristic views. Each view is represented by a set of "eigenfaces" which are the significant eigenvectors (principal components) of the set of known faces. They form a holistic representation and do not necessarily correspond to individual features such as eyes, ears, and noses. This approach provides for the ability to learn and later recognize new faces in an unsupervised manner. In addition to face recognition, I explore other visual behaviors in the domain of human-computer interaction. Thesis Supervisor: Alex P. Pentland Associate Professor, MIT Media Laboratory

Recognition in Face Space

Abstract: Individual facial features such as the eyes or nose may not be as important to human face recognition as the overall pattern capturing a more holistic encoding of the face. This paper describes " face space" a subspace of the space of all possible images which can be described as linear combinations of a small number of characteristic face-like images. The construction of face space and its use in the detection and identification of faces is explained in the context of a working face recognition system. The effects of illumination changes scale orientation and the image background are discussed.© (1991) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

'The digital doorkeeper'-automatic face recognition with the computer

Abstract: The problem of automatic face recognition is investigated. A multiresolution representation of the scene is scanned with a matched filter based on local orientation for the reliable localization of human faces. For the identification of the faces, two complementary strategies are used. At low resolution, the three most important features of a face (head, eye pairs, and nose/mouth/chin) are compared with the contents of a database. At high resolution, the precise location of several landmark features is determined, and this geometrical description is used for comparisons in a 62-dimensional vector space. These algorithms were used in an experimental access control system-the digital doorkeeper-to investigate its performance under realistic conditions. Without screening for spectacles, beards. changing hairstyle, etc. a recognition rate of 90% among known persons was achieved. At a recognition rate of 60% for known persons, less than 3% of unknown persons were wrongly admitted. >

Minimum resolution for human face detection and identification

Abstract: Our goal is to build an automated system for face recognition. Such a system for a realistic application is likely to have thousands, possibly miffions of faces. Hence, it is essential to have a compact representation for a face. So an important issue is the minimum spatial and grayscale resolutions necessary for a pattern to be detected as a face and then identified. Several experiments were performed to estimate these limits using a collection of 64 faces imaged under very different conditions. All experiments were performed using human observers. The results indicate that there is enough information in 32 x32 x 4bpp images for human eyes to detect and identify the faces. Thus an automated system could represent a face using only 512 bytes.

Image based target maneuver detection

Abstract: A noncooperative target tracking application with a forward-looking infrared (FLIR) sensor is considered. Results are reported that demonstrate the feasibility under realistic conditions of target maneuver detection using a postprocessing of target attributes (e.g. the number of pixels or width) extracted from a sequence of FLIR images. Synthetic realistic FLIR images were generated for an air encounter scenario using a CAD facet model of the target to provide the scene geometry and a temperature knowledge base plus an atmospheric transmission code to compute the radiometries. >

GaAs Opto-Electronic Integrated Circuits for High Speed Optical Communications

Abstract: Opto-Electronic-Integrated-Circuits (OEICs) fabricated using GaAs circuit technologies are particularly useful in computer applications. A number of such OEIC chip are described illustrating the versitility of this technology.

Model Based Recognition Of Underwater Objects

Abstract: Automatic recognition of objects is a challenging and difficult task that has shown to be elusive during the past two decades. What has emerged from the years of experimentation in building recognition systems based on passive infrared(1R) and active millimeter wave (MMW)radar sensors is the new model bused paradigm . The application and appreciation of this new concept in the field of underwater sensing, however, has been non-forthcoming. In this paper I attempt to discus the basics of model based target recognition and show how this paradigm can be used in automatic classification of underwater data using the SeaMARC imagery as an example.