Computer graphics processing and selective visual display system – Computer graphics processing – Graphic manipulation
Reexamination Certificate
1998-04-29
2004-05-04
Luu, Matthew (Department: 2672)
Computer graphics processing and selective visual display system
Computer graphics processing
Graphic manipulation
C345S646000
Reexamination Certificate
active
06731302
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to a system for creating facial images. It is applicable to the creation of composite pictures, specifically to systems used in law enforcement, artistic creations, recreation and education.
BACKGROUND OF THE INVENTION
Human facial characteristics can convey a large amount of information and are a very common means for quickly identifying individuals. These facial characteristics are particularly useful in the area of law enforcement where composite facial images are often used to identify criminal suspects. They are also useful to create portraits in cases where no picture of the person to be graphically described is available.
Traditionally, these composite pictures are created by graphic artists who confer with an eyewitness. Through lengthy trial and error sessions, attempts are made to create a composite picture of a suspect.
The traditional method described above presents two problems. The first is that the accuracy of a composite picture is related to the graphic artist's ability to translate the description of an eyewitness into an image. During the process, the memory of an eyewitness may be inadvertently altered by comments from the artist or from other third parties resulting in a less than accurate portrait of the suspect. The second problem with the traditional method is the high labor costs associated with the graphic artists themselves. Because police departments, in particular those in large urban centers, are likely to require several of such composite pictures, the associated labor costs are significant. Consequently, in order to overcome these problems, numerous attempts have been made to permit the witnesses to create the facial image with little or no interference from third parties. In addition to limiting the influence of a third party on the witness, the labor costs associated with the creation of composite pictures are also reduced.
One early technique for synthesizing single images of faces involved horizontally dividing the image of a face into bands for different features of the face such as hair, eyes, nose, mouth, and chin, respectively. Paper strips containing exemplary features could then be combined to form a composite drawing of a face.
One disadvantage of these systems is the presence of discontinuities between the various bands. For example, the size of the eyes may be disproportionate with that of the head or mouth. Furthermore, the skin tone between different bands may not be identical. These discontinuities often create and image which is cartoon-like and which bears little resemblance to the real life image the user of the system may wish to create.
Other techniques involve selecting individual components and combining them on a pre-selected face. In a typical interaction, the user first selects the shape of the face then eyes, nose, mouth and other components and combines them to form a facial image. Many variations on this theme can be used as described in Kakiyama et al. U.S. Pat. No. 5,600,767, Yoshino et al. U.S. Pat. No. 5,644,690, Sato et al. U.S. Pat. No. 5,537,662 and Belfer et al. U.S. Pat. No. 5,649,086 whose contents are hereby incorporated by reference. For example, the Sato et al. Patent, entitled Electronic Montage composing apparatus, describes a system for creating a montage image of a face using a plurality of basic parts stored in a library. In a typical interaction, the user selects a basic face from a plurality of basic faces and replaces the parts of the face in order to obtain the desired final result. Only the parts that are different from the ones desired need to be modified. This allows the user to quickly obtain a facial image without requiring the selection of each individual part. The Belfer et al. Patent describes another variation on this combination of individual components. This invention includes a database of exemplar images, or model components, which are stored and indexed according to various characteristics. Images are created by interpolating among the exemplar images in the database based on user selection of parameter values. Examples of high level parameters are gender and expression (surprise, anger). Examples of low level parameters are nose length, nostril width, hair thickness and so on. This system allows storing a reduced number of components by producing combinations of the components through interpolation. By allowing the user to vary parameters across a range of extreme values, facial images with the desired appearance can be created.
Although the use of individual components allows the user to position them at the desired location on the face, the proportions of the components often remains unsatisfactory. Furthermore, discontinuities remain and are often significant particularly when the facial images created comprise components selected from various genders and races. Furthermore in the case of the Belfer et al. Patent, tuning the parameters to obtain the desired appearance may be lengthy and difficult to master for the inexperienced user requiring the presence of an experienced technician.
Thus, there exists a need in the industry to refine the process of creating facial images such as to obtain a better composite picture system particularly applicable to law enforcement, artistic creations, recreational use and education.
OBJECTS AND STATEMENT OF THE INVENTION
An object of the invention is to provide a novel method and an apparatus for creating facial images.
Another object of this invention is to provide a computer readable storage medium containing a novel program element that directs a computer to perform the creation of a facial image.
Another object of this invention is to provide a computer readable storage medium containing novel library of basic morphological elements.
Another object of this invention is to provide a novel method for creating a library of basic morphological elements.
As embodied and broadly described herein the invention provides an apparatus for generating a composite facial image, said apparatus comprising:
machine-readable storage medium for storing a set of data elements, said data elements being representative of basic morphological elements, said data elements being in a format such that graphical representations of the basic morphological elements associated with the data elements are proportionate to one another;
an input for receiving user data indicative of a plurality of data elements;
processing means in operative relationship with said machine-readable storage medium for processing the plurality of data elements to provide a digital representation of a facial image resulting from a combination of the basic morphological elements associated with the plurality of data elements.
For the purpose of the specification, the expressions “morphological element”, “morphological component” and “basic morphological elements” are used to describe a part of a specific facial image. Examples of morphological elements are noses, eyes, mouths and eyebrows. In the preferred embodiment, morphological elements are grouped into morphological classes. For example, all nose elements are grouped into the “NOSE” morphological class and all the eye elements are grouped in the “EYE” morphological class. Optionally, the classes are divided into subclasses to further specify the types of elements stored. For example, the morphological class “MOUTH” may be subdivided into “thick lips”, “narrow”, “thin lips” subclasses.
For the purpose of the specification, the expression “calibrating” is used to designate the process of modifying the proportions of a set of distinct morphological components in order to improve their interaction. For example, the basic morphological elements in the library of basic morphological elements are calibrated to produce a facial image where the various parts can be connected with reduced discontinuities at their juncture. This is particularly useful when the volumes of the heads of different individuals from which the basic morphological elements were taken vary significantly.
For the purpose of
Fleshner & Kim LLP
Havan Thu-Thao
IQ Biometrix, Inc.
Luu Matthew
LandOfFree
Method and apparatus for creating facial images does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for creating facial images, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for creating facial images will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3271229