Computer graphics processing and selective visual display system – Computer graphics processing – Attributes
Reexamination Certificate
2001-08-15
2004-01-13
Bella, Matthew C. (Department: 2676)
Computer graphics processing and selective visual display system
Computer graphics processing
Attributes
C345S589000, C345S629000, C345S582000, C345S419000, C345S667000, C345S443000
Reexamination Certificate
active
06677956
ABSTRACT:
FIELD OF THE INVENTION
The invention relates generally to rendering graphics, and more particularly to cross-fading overlapping images.
BACKGROUND OF THE INVENTION
Computer graphics systems can acquire or synthesize images of real or imaginary objects and scenes, and then reproduce these in a virtual world. More recently, computer systems have also attempted to do the reverse—to “insert” computer graphics images into the real world. Primarily, this is done indirectly for special effects in movies, and for real-time augmented reality. Most recently, there is a trend to use light projectors to render imagery directly in real physical environments.
Despite the many advances in computer graphics, the computer has yet to replace the actual material experience of physical shape and spatial relationships. Designers, such as architects, urban planners, automotive engineers, artists and animators still resort to sculpting physical models before the design is finalized. One reason for this is that the human interface to a physical model is totally intuitive. There are no controls to manipulate, or displays to look through or wear. Instead, the model can be viewed from many perspectives while gazing generally or focusing on interesting components, all at very high visual, spatial, and temporal fidelity.
When an object or scene is illuminated by a neutral (white) light, it is perceived according to the particular wavelengths of the light reflected by its surface. Because the attributes of the surface are dependent only on the spectrum of the perceived light, many attributes of objects can effectively be simulated by incorporating the object's attributes into the light source to achieve an equivalent effect on a neutral object. Thus, even non-realistic appearances can be visualized.
A rotating movie camera has been used to acquire a film of a living room, replete with furniture, and people. The room and furniture were then painted a neutral white, and the film was projected back onto the walls and furniture using a rotating projector that was precisely registered with the original camera, see Naimark, “Displacements,” Exhibit at the San Francisco Museum of Modern Art, San Francisco, Calif. 1984. This crucial co-location of the acquiring camera and displaying projector is common to most systems that use pre-recorded images, or image sequences to illuminate physical objects.
A projector and fiber-optic bundle have been used to animate the head of a fictional fortune teller inside a real crystal ball, see U.S. Pat. No. 4,978,216 “Figure with back projected image using fiber optics” Liljegren, et al., Dec. 18, 1990. Slides of modified photographs augmented with fine details have been used with very bright projectors to render imagery on a very large architectural scale. A well known modern realization of this idea is Le Son et Lumière on Château de Blois in the Loire Valley of France. In addition, this medium is now being used elsewhere around the world to illuminate large scale structures such as bridges.
All these systems render compelling visualizations. However, cumbersome alignment processes can take several hours, even for a single projector. The “Luminous Room” project treats a co-located camera-projector pair as an I/O bulb to sense and project imagery onto flat surfaces in the real physical surroundings of a room or a designated workspace, see Underkoffler et al. “
Emancipated pixels: Real-world graphics in the luminous room,” SIGGRAPH '
99, pp. 385-392, 1999. Their main focus is interaction with the information via luminous and tangible interfaces. They recognized co-planar 2D physical objects, tracked the 2D positions and orientations in the plane, and projected light from overhead to reproduce the appropriate sunlight shadows.
In the “Facade” project, a sparse set of photographs was used to model and render architectural monuments, see Debevec et al. “
Modeling and Rendering Architecture from Photographs
,” SIGGRAPH '96, August 1996. Their main problems were related to occlusion, sampling, and blending issues that arise when re-projecting images onto geometric models. They addressed these problems with computer images and analytic models.
It would be useful to have a graphics system that can project images onto real three-dimensional objects or structures. It should also be possible to fit the images for any viewing orientation. In addition, it should be possible to change the appearance of illuminated objects at will. Finally, it should be possible to seamlessly fit images from multiple projectors onto a single complex object so that it can be viewed from any direction.
SUMMARY OF THE INVENTION
The invention provides a novel mode for visualizing three-dimensional computer graphics. The invention can synthesize not just textures, i.e., diffuse component of images, but also reproduce virtually any bi-directional reflectance distribution function (BRDF) using a radiance adjustment procedure. The invention also determines weighted pixel intensities across transitions of multiple overlapping images in the presence of depth discontinuities.
The present paradigm for 3D computer graphics uses multiple projectors to graphically animate physical objects or structures in the real world. The concept is to replace a physical object having inherent color, texture, and material properties with a neutral object illuminated by projected imagery, reproducing the original or alternative appearance directly on the object. In this approach, the desired visual properties of the object are effectively “lifted” into the projector.
Specifically, a computer implemented method cross-fades intensities of a plurality of overlapping images by identifying pixels in a target image that are only produced by a first source image. The weights of all the corresponding pixels in the first source image are set to one. Pixels in a second images contributing to the target image are similarly identified and set to one. The weight of each remaining pixel in the first and second images is set inversely proportional to a distance to a nearest pixel having a weight of one. The source images are then projected to produce the target image.
REFERENCES:
patent: 4503427 (1985-03-01), Iida
patent: 5903279 (1999-05-01), Lee et al.
patent: 5905499 (1999-05-01), McDowall et al.
patent: 6285370 (2001-09-01), McDowall et al.
patent: 2001/0017649 (2001-08-01), Yaron
Debevec et al, “Efficient View-Dependent Image-Based Rendering with Projective Texture-Mapping”; Proceedings of 9th Eurographics Workshop on Rendering, 1998.
Low Kok-Lim
Raskar Ramesh
Welch Gregory F.
Bella Matthew C.
Brinkman Dirk
Curtin Andrew
Mitsubishi Electric Research Laboratories Inc.
Rahmjoo Mike
LandOfFree
Method for cross-fading intensities of multiple images of a... 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 for cross-fading intensities of multiple images of a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for cross-fading intensities of multiple images of a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3221395