Computer graphics processing and selective visual display system – Computer graphics processing – Three-dimension
Reexamination Certificate
2000-10-31
2004-08-17
Sealey, Lance W. (Department: 2671)
Computer graphics processing and selective visual display system
Computer graphics processing
Three-dimension
C348S169000, C342S357490, C342S357490, C342S357490
Reexamination Certificate
active
06778171
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates in general to object tracking systems. More particularly, it relates to a method and apparatus for object tracking that involves using a three-dimensional (“3D”) graphics pipeline in a novel manner to facilitate manipulation of digital images.
2. Description of the Related Art
The task of tracking objects of interest has applications in many problem domains. In image processing, for example, the tracking of objects through time is a very common task. In the field of computer vision, the tracking of objects is a key component, e.g., for accomplishing robotic tasks in an environment that includes independently moving objects or for accomplishing robotic interception of fast-maneuvering objects. Object (also known as “target”) recognition and tracking is an important research area in pattern recognition. Industrial and commercial applications, e.g., those in which vision-based control of motion is feasible only if vision provides reliable control signals, involve object tracking. Moreover, with the continuous advances in wireless communications as well as positioning technologies, tracking the changing positions of objects is becoming increasingly feasible.
The methodologies in tracking moving objects are influenced by the natural environment and motion characteristics. In the case of image processing, the tracking of objects through time normally requires the definition of the object boundaries in order to describe its motion. But such an approach does not always give sufficient information about the position of specific points located on the surface of the object, and this is especially a problem when the shape of the object also changes. In some research, e.g., in the biomedical field, one approach is to restrict the assessment to the motion and morphology of line-like 3D-structures (lines or tubes) by tracking the positions of small segments of such structures in two orthogonal projections, as disclosed in M. Schrijver, J. W. Homan & C. H. Slump,
Estimating The
3
D Motion And Morphology Of Line
-
Like Structures From Orthogonal Projections
, PRoRISC/IEEE 10
TH
Annual Workshop on ′Circuits Systems and Signal Processing (November 1999). Another approach makes use of a prediction model for moving object velocity and location estimation derived from Bayesian theory, as disclosed in A. G. Bors & I. Pitas,
Prediction And Tracking Of Moving Objects In Image Sequences
, IEEE Transaction on Image Processing, vol. 9 (no. 8), pp. 1441-45 (August 2000).
In pattern recognition, some systems for target recognition and tracking are based on a single sensor, e.g., radar or infrared image sensor. Such single-sensor systems have their limitations. Other target recognition and tracking systems are based on data fusion of radar/infrared image sensors. Data fusion—at a characteristic level—can combine characteristics from different sensors to improve the ability of object recognition. Whereas, data fusion at the decision level can improve the reliability of object recognition.
Object location and tracking systems are particularly useful in asset management systems where continuous identification of the locations of a variety of objects is desired. An invention directed to an RF-energy based object tracking system, using correlation processing to identify a first-to-arrive burst from a tag, and a multilateration algorithm, which is able to identify the locations of objects, is disclosed in U.S. Pat. No. 5,920,287, issued Jul. 6, 1999 to Belcher et al.
A method of tracking a target object using a computer system comprising the steps of capturing a first image using an imaging device, wherein the first image corresponds to the target object, and generating a second image from the first image, wherein the second image also corresponds to the target object, and using a selection signal that corresponds to a selection of the second image is disclosed in U.S. Pat. No. 6,061,055, issued May 9, 2000 to Marks. A tracking strategy that can track a person or other object within a fixed environment using a pan, tilt, and zoom camera with the help of a pre-recorded image database is disclosed in Yiming Ye et al.,
Tracking A Person With Pre
-
Recorded Image Database And A Pan, Tilt, And Zoom Camera
, Machine Vision and Applications, vol. 12 (no. 1), pp. 32-43 (Springer-Verlag, 2000). This tracking strategy, however, requires defining a set of camera states to survey the environment for the target and, during tracking, camera movements are restricted to these states.
A system for monitoring and tracking objects as they traverse a predefined space is disclosed in U.S. Pat. No. 5,973,732, issued Oct. 26, 1999 to Guthrie. Other systems have been used which utilize video cameras and processing devices for tracking objects and individuals. Some example devices are shown in U.S. Pat. No. 5,243,418, issued Sep. 7, 1993 to Kuno et al. A system for detecting and monitoring an area is disclosed in U.S. Pat. No. 5,097,328, issued Mar. 17, 1992 to Boyette. This system is tailored to monitor customer service windows at a bank and uses a combination of blob analysis and motion analysis to determine if a server is on-duty and present at his station, and if he is serving a customer. A system for monitoring the flow of traffic through a designated area, in which a video traffic monitor counts the detected people and records the count according to the direction of movement of the people, is disclosed in U.S. Pat. No. 5,465,115, issued Nov. 7, 1995 to Conrad et al. In this system, an imaginary row of gates is defined, and images filling one or more of these gates are monitored.
As demonstrated in the prior art, many tracking algorithms have been proposed which work well under some conditions, but have easily defined conditions for which they fail. One important problem that has yet to be completely solved is recognizing (tracking) select objects in a scene on a display screen. In this problem space, regional and structural features of objects include shape, texture, color and edges. Existing methods typically require complex and error-prone image-recognition approaches and/or active intervention by an editor, i.e., one editor per object. Such tracking methods often employ artificial intelligence in seeking to do image recognition. When it happens that a moving object has distinct color and a smooth path, a projection of the direction by the artificial intelligence can be used to track the object, within limits. However, if a high-powered zoom lenses is used to follow an object that exhibits erratic motion, then tracking becomes problematic. Also, sudden changes in color can disrupt object tracking. Such limitations in the existing methods effectively preclude the use of real-time object hyperlinking and tracking methods in real-time broadcasts of sports events or other high-motion events, such as car racing.
What is clearly needed is a method of tracking objects that allows objects to be tracked across multiple scene changes, with different camera positions, without losing track of the selected object.
SUMMARY OF THE INVENTION
Accordingly, this invention provides a method of tracking objects that allows objects to be tracked across multiple scene changes, with different camera positions, without losing track of the selected object.
In one embodiment, a method of tracking an object using a computer, a display device, a camera, and a camera tracking device, the computer being coupled to the display device, the camera and the camera tracking device is disclosed. The method includes: A first image from within a field-of-view of the camera is captured. The first image, which includes an actual object with a tracking device, is displayed on the display device. Information about the tracking device's location is received. The information is used to create a virtual world reflecting the actual object's position within the field-of-view of the camera as a shape in the virtual world. Information about the camera tracking device is received. A v
Depke Robert J.
Eagle New Media Investments, LLC
Holland & Knight LLP
Sealey Lance W.
LandOfFree
Real world/virtual world correlation system using 3D... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Real world/virtual world correlation system using 3D..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Real world/virtual world correlation system using 3D... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3318978