Graphic scene animation data signal with quantization...

Details Graphic scene animation data signal with quantization... Graphic scene animation data signal with quantization...

2000-03-07

2003-04-29

Zimmerman, Mark (Department: 2671)

Computer graphics processing and selective visual display system

Computer graphics processing

Animation

C345S428000

Reexamination Certificate

active

06556207

ABSTRACT:

The domain of the invention is coding of data representative of animated graphic scenes in order to reproduce them, for example on a multimedia terminal. More precisely the invention relates to the quantization of this type of data, so that they can be stored and/or transmitted.
For the purposes of this description, a graphic scene means the layout of graphic, video and image objects in time and in space. This type of graphic scene may be in two or three dimensions and may contain various types of graphic primitives.
The invention may be used in applications in all cases in which elements of a graphic scene have to be coded. In particular, it is applicable to scenes using the known VRLM and BIFS scene description formats developed in MPEG-4.
Standard ISO/IEC DIS 14772-1 describes the VRML 2.0 format. The MPEG-4 standardization group defined the scene description format called BIFS (Binary Format for Scene) which is based on VRML 2.0. The BIFS format is described particularly in “The MPEG-4 Systems Verification Model” (ISO/IEC JTC1/SC29/WG 11-N1693, MPEG 97, April 1997).
The purpose of this scene description format is to describe space/time relations between the various graphic objects in a scene. It does this by defining a number of nodes or objects representing all graphic primitives that are to be represented. Each of these nodes includes predefined fields that represent the characteristics of these nodes.
In other words, the BIFS format transmits a scene structure in the form of a parametric description, or a script.
BIFS type graphic scenes may be used for multimedia viewing (remote teaching, remote purchasing, remote work services), 3D games, advanced navigation interfaces in the services. In particular, the invention may be useful in the following applications
viewing VRML 2.0 type graphic scenes on an “Internet” type network, in which a compact data display format has to be defined in order to reduce transmission times,
the storage of this type of scene on media such as CD-Roms.
A binary scene display format already known for the VRML 2.0 standard is described in the document mentioned above. In order to quantize the parameters, the author proposes to send quantization parameters in the form of a node. However, the chosen quantization parameters are only applicable to a very limited number of fields. This has the effect of imposing the transmission of a large number of fields in an unquantized manner. Obviously, this limits the efficiency of the digital compression of data.
One particular purpose of the invention is that it overcomes these disadvantages in the state of the art.
More precisely, one purpose of the invention is to supply a data signal, and a process and device for using this signal, in order to significantly reduce the data necessary for coding and therefore the transmission and/or storage of animated graphic scenes, particularly in standards such as VRML and MPEG-4.
Another purpose of the invention is to provide a coding technique for data representative of graphic scenes that enable transmission of scenes at low throughput on a network, and reconstruction of these scenes in terminals that do not necessitate major hardware or software means.
Another purpose of the invention is to provide such a technique that will be efficient with any type of scene and any type of element forming part of this scene, without the need for specific (in other words in an unquantized manner) processing of some elements.
These purposes, and other purposes that will be described later, are achieved according to the invention by using a data signal for animation of a graphic scene to be used by means for constructing images that can be displayed on at least one screen, a signal in which the said scene is described in the form of a set of animation objects, each of which is associated with at least one characterization field defining a parameter of the said object, and comprising at least one quantization object, for which the characterization fields define the quantization rules for the characterization fields of the said animation objects, each applicable to at least two distinct characterization fields, such that most or all of the characterization fields of the said animation objects with a numeric value can be quantized.
Thus, a limited number of quantization types or rules can be defined to cover all possible cases.
Furthermore, it is proposed to transmit quantization parameters as an object or node. The advantage of transmitting these parameters as a node is to benefit from all functions related to transmission of nodes in a scene description stream:
declaration of a node in the scene is equivalent to declaring all its fields, with predefined default values;
declaration of any field is a means of redefining a value other than the default value of this field;
this node can be identified and then reused whenever necessary in the scene, simply by indicating its identifier.
Preferably, this type of quantization object includes at least one of the quantization rules belonging to the group comprising:
quantization rule for a three-dimensional position;
quantization rule for a two-dimensional position;
quantization rule for a color;
quantization rule for a texture;
quantization rule for an angle;
quantization rule for a scale change;
quantization rule for an animation key;
quantization rule for a normal.
Advantageously, each of the said quantization objects comprises a Boolean range field, indicating:
for a first value, that the quantization object is only applicable to the next object;
for the second value, that the quantization object is applicable to all subsequent objects, until a new quantization object is found.
The number of necessary data can thus be further limited.
According to one preferred embodiment, at least some of the items of information belonging to the group comprising the following may be produced for each rule for quantization of a parameter type:
a flag stating whether or not the quantization is to be used;
a minimum value (min) of the said parameter;
a maximum value (max) of the said parameter;
a number of bits (Nb) assigned to quantization of the said parameters.
The invention also relates to a coding process capable of producing this type of signal. This process comprises in particular:
a definition step, outputting at least one quantization object in which the characterization fields define rules for quantization of characterization fields of the said animation objects, each applicable to at least two distinct characterization fields, such that most or all of the characterization fields with a numeric value of the said animation object may be quantized; and
a quantization step, quantizing data specific to each of the said characterization fields of the said animation objects as a function of the said quantization rules.
Finally, the invention also relates to a device for the construction of images that can receive this signal. In particular, it comprises:
means of reception of at least one quantization object, for which the characterization fields define the rules for quantization of the characterization fields of the said animation objects, each applicable to at least two distinct characterization fields, such that most or all characterization fields of the said animation objects with a numeric value may be quantized; and
inverse quantization means for data specific to each of the said characterization fields of the said animation objects, as a function of the said quantization rules.


REFERENCES:
patent: 5315692 (1994-05-01), Hansen et al.
patent: 5742289 (1998-04-01), Naylor et al.
patent: 5745101 (1998-04-01), Yamamoto et al.
patent: 5818434 (1998-10-01), Yamamoto et al.
Rose, K. Et al.: “Entropy—Constrained Tree—Structured Vector Quantizer Design”; IEEE Transactions on Image Processing; vol. 5, No. 2; 1996; pp. 393-398; XP-000678941.
Koski, A.: “Primitive Coding Structural ECG Features”; Pattern Recognition Letters; vol. 17, No. 11, Sep. 16, 1996, pp. 1215-1222; XP-000639569.
Jasinschi, R. S. Et al.; “Video Compression Via Constructs”; Proceedings

Affiliated with

Also associated with

Rating

Graphic scene animation data signal with quantization... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Graphic scene animation data signal with quantization..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Graphic scene animation data signal with quantization... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3001108