Image analysis – Image compression or coding – Shape – icon – or feature-based compression
Patent
1997-06-27
1999-09-28
Couso, Jose L.
Image analysis
Image compression or coding
Shape, icon, or feature-based compression
382286, G06K 900
Patent
active
059601183
DESCRIPTION:
BRIEF SUMMARY
FIELD OF THE INVENTION
This invention relates to apparatus and method for representing 2-dimensional and 3-dimensional scenes and images in a compressed form, particularly, but not exclusively, for the purpose of storing and/or transmitting the compressed data and subsequently reconstructing the picture or the 3D-scene, as seen from any prescribed point, in a manner as faithful as possible.
BACKGROUND OF THE INVENTION
The representation of various objects by data compression is a problem with which the art has been increasingly occupied in recent times. As far as the usual images and videosequences are concerned, a background of the present invention is described in the Israeli patent application IL 103389 (the priority of which is claimed in U.S. Pat. No. 5,510,838). Representation of three-dimensional scenes and objects, which becomes increasingly important in recent times, requires imcomparably bigger amounts of information. Also fast representation and rendering of three-dimensional scenes as seen from an interactively prescribed point, presents difficult computational problems. The invention presents a method for an automatic capturing and highly compressed representation of 2D and 3D scenes and images with a subsequent fast rendering.
SUMMARY OF THE INVENTION
A. Main applications of the invention
A 3D-itinerary is a compact representation of a chain of 3D-scenes along a certain camera trajectory. The user can interactively create (in real time on a PC or a comparable platform) a photorealistic quality images of the represented scenes, as seen from any prescribed point in a vicinity of the initial trajectory. In particular, the user can "fly" in real time inside the represented space.
The product consists of two parts: of 3D-iteneraries. An input for a preparation of a 3D-itinerary is a usual videosequence, showing the desired 3D-scenes from a certain camera path. Then the package completely automatically creates the corresponding 3D-itinerary (in minutes of a software processing for one scene). "traveling" along the 3D-itinerary.
Our 3D-iteneraries are strongly compressed, and the data volume of a complete 3D-representation of a certain scene (which allows one to see if from any prescribed point), is comparable with the data volume of a single (noncompressed) 2D-image of the same scene.
The main proposed applications include computer assisted training, advertising, computer games and other multimedia applications.
It is important to stress that the 2D-images, reconstructed from a 3D-itinerary, faithfully represent the scene, so creation of 3D-iteneraries can be alternatively called "3D-compression". In applications, where the fidelity to the original is less important, while the high quality of the images must be, of course, preserved (like in computer games imaginary) much more compact virtual 3D-images and virtual 3D-iteneraries can be used (see below).
A virtual 3d-image is a highly compressed representation of a virtual 3D-scene, created from one or several high quality still images by special interactive tools. The end-user interactively produces in real time photorealistic quality images of this scene from any desired point. These images are not authentic to the real views of a certain 3D-scene, but they provide a complete visual illusion of the required motion inside the scene.
The product consists of two parts: consists of one or several still images. The toolkit allows the user to interactively create a 3D-structure on these images, to superimpose them, to supply "invisible parts" etc. In many aspects our tools are similar (and in fact, include) the standard toll used in Pre-Press and Desk-Top Publishing image processing. used for 3D-iteneraries.
Virtual 3D-images are compressed to 1/10 up to 1/50 of the volume of the usual still image. They can be combined into virtual 3D-iteneraries, similar to the complete 3D-iteneraries, described above. However, virtual 3D-iteneraries are compressed to a much smaller data volume, they do not require detailed videosequences of the scene for their
REFERENCES:
patent: 4956869 (1990-09-01), Miyatake et al.
patent: 4969201 (1990-11-01), Takasaki et al.
patent: 5020121 (1991-05-01), Rosenberg
patent: 5036544 (1991-07-01), Sakave et al.
patent: 5048112 (1991-09-01), Alves et al.
patent: 5050222 (1991-09-01), Lee
patent: 5086482 (1992-02-01), Kumagai
patent: 5231678 (1993-07-01), Takatori et al.
patent: 5243665 (1993-09-01), Maney et l.
patent: 5249035 (1993-09-01), Yamanaka
patent: 5410643 (1995-04-01), Yomdin et al.
patent: 5420942 (1995-05-01), Levit
patent: 5510838 (1996-04-01), Yomdin et al.
Briskin Miriam
Elihai Yoram
Yomdin Yosef
Couso Jose L.
Tadayon Bijan
LandOfFree
Method for 2D and 3D images capturing, representation, processin 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 2D and 3D images capturing, representation, processin, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for 2D and 3D images capturing, representation, processin will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-713077