Statistically comparing and matching plural sets of digital data

• US 20060083405A1
• Filed: 12/07/2005
• Published: 04/20/2006
• Est. Priority Date: 02/18/2000
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Szeliski, Richard S., Wilt, Nicholas P.
• Granted Patent: US 7,574,017 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors
• CPC Class Codes: G06V 10/94 Hardware or software architectures specially adapted for image or video understanding

Abstract

The present invention is embodied in a system and method for statistically comparing a first set of digital data to at least a second set of digital data and matching the first set of digital data to appropriately corresponding portions of the second set of digital data. The first or the second set of digital data can be transformed during statistical analysis to enhance statistical analysis of the digital data.

Claims

1. A method for comparing and matching a first set of digital data to at least a second set of digital data, comprising:

using a computer graphics card for raster transforming at least one of the first set of digital data and the second set of digital data said raster transformations including combinations of rotations, scales, perspective transforms, and translations;

depending upon the results of a pixel acceptance test performed by an acceptance tester included in the computer graphics card, performing a statistical comparison between at least part of the first set of digital data and at least part of the second set of digital data using a statistical processor included in the computer graphics card;

wherein the statistical processor uses pixel alpha values to weight any of the first set of digital data and at the second set of digital data; and

wherein the statistical comparison includes statistically comparing and matching the alpha weighted and raster transformed sets of digital data to appropriately corresponding portions of each other using the statistical processor.

2. The method of claim 1, further comprising using the statistical processor for analyzing the statistical comparisons and using the computer graphics card for generating new transformations for matching the sets of data.

3. The method of claim 1, further comprising using the statistical processor for statistically comparing the raster transformed sets of digital data until a match or non-match between the first and second sets of data is achieved.

4. The method of claim 1, wherein the raster transforming comprises raster transforming at least one of the first or the second set of digital data and computing statistics on the transformation.

5. The method of claim 4, wherein statistically comparing and matching comprises analyzing the computed statistics of the transformation and calculating new and different transformations on the digital data.

6. A method for comparing and matching a first set of digital data to at least a second set of digital data, comprising:

loading at least one of the first and second sets of digital data into a first memory device;

using a 3D graphics rendering device for rendering model transformations and accumulating statistics of the loaded digital data, said 3D graphics rendering device modified to include a statistical processor, and said model transformations including combinations of rotations, scales, perspective transforms, and translations;

weighting the accumulated statistics using pixel alpha values associated with any of the first and second sets of digital data;

adjusting the model transformations based on the weighted accumulated statistics; and

statistically comparing and matching the model transformations of the loaded set of digital data to appropriately corresponding portions of the other set of digital data.

7. The method of claim 6, further comprising statistically comparing the sets of digital data until a match or non-match between the first and second sets of data is achieved.

8. The method of claim 6, wherein adjusting the model transformations comprises analyzing the statistical comparisons and generating new transformations for matching the sets of data.

9-20. -20. (canceled)