Method of generalized content-scalable shape representation and coding

• US 6,011,872 A
• Filed: 05/14/1997
• Issued: 01/04/2000
• Est. Priority Date: 11/08/1996
• Status: Expired due to Term

• Current Assignee: Rakuten, Inc. (Rakuten Group, Inc.)
• Original Assignee: Sharp Laboratories of America Incorporated (Hon Hai Precision Industry Co., Ltd.)
• Inventors: Qian, Richard J., Sezan, M. Ibrahim
• Primary Examiner(s): Couso, Jose L.
• Application Number:
  US08/856,387
• Time in Patent Office: 965 Days
• Field of Search: 382/232, 382/233, 382/236, 382/238, 382/239, 382/241, 382/242, 382/243, 382/245, 382/246, 382/250, 382/258, 348/384, 348/390, 348/400-402, 348/409, 348/412, 348/413, 348/416, 348/420, 348/428
• US Class Current: 382/243
• CPC Class Codes: G06T 3/40 Scaling of whole images or parts thereof, e.g. expanding or contracting
• CPC Class Codes: G06T 9/20 Contour coding, e.g. using detection of edges
• CPC Class Codes: H04N 19/20 using video object coding
• CPC Class Codes: H04N 19/39 involving multiple description coding [MDC], i.e. with separate layers being structured as independently decodable descriptions of input picture data

Abstract

A method for content-scalable shape representation is described. In content-scalable shape representation, the more important parts of the image are encoded using a larger number of bits within a single hierarchical layer, or within one or more layers of multiple hierarchical layers. This representation may be used with the framework of a hierarchical vertex-based shape representation technique. The method described herein provides for the representation of shapes either by selected verticles along a contour, or by binary bit maps. The resulting representation facilitates content-scalable coding.

Claims

1. A method of content-scalable, shape representation and coding for use on an image object, comprising:

forming a binary bitmap of the shape of an image object;

dividing the binary bitmap into a macroblock array;

defining important segments of the shape of an image object and defining non-important segments of the shape of the image object, wherein an important segment is a segment that surrounds a portion of the image object to be represented and coded with greater accuracy than an unimportant segment;

coding an important segment with more bits than are used to code a non-important segment;

forming a multi-resolution representation of each macroblock, wherein the multi-resolution representations range from a coarse resolution through successive layers of finer resolutions, and wherein the number of finer resolution representations is controlled for each macroblock on the basis of the visual importance of the region represented by the macroblock.

2. The method of claim 1 which includes coding the important parts of a shape using relatively more hierarchical layers of bit maps at different resolutions for each macroblock.

3. The method of claim 1 which includes, for each macroblock, selecting a total number of layers;

and selecting a saliency parameter for each layer and for each macroblock.

4. The method of claim 3 which includes associating a scalability indicator with each macroblock.

5. The method of claim 4 which includes coding unimportant macroblocks only in a single layer.

9. The method of claim 1 wherein the important segment is defined as that segment surrounding an object selected by the user.

6. A method of content-scalable, shape representation and coding for use on an image object, comprising:

representing the shape of the image object using plural hierarchical layers, wherein each layer represents the shape of the image object at a different level of accuracy;

determining important segments and non-important segments of an image object shape from the image object, wherein an important segment is a segment that surrounds a portion of the image object to be represented and coded with the highest accuracy;

defining a contour of interest about the periphery of the important segment and using a relatively larger number of vertices to define the contour of interest;

selecting, for important segments of an image object shape, a total number of layers for each segment, and selecting, for non-important segments, a lesser number of layers than for the important segments;

selecting a saliency parameter for each layer and for each segment; and

coding the important segment of the image object with more bits allocated thereto than to the non-important segments of the image object.

7. The method of claim 6 which includes dividing the image object shape into macroblocks;

and associating a scalability indicator with each macroblock.

8. The method of claim 7 which includes coding unimportant macroblocks only in a single layer.

10. The method of claim 6 wherein the important segment is defined as that segment surrounding a head.