Compressing image-based data using luminance

• US 8,594,441 B1
• Filed: 09/12/2006
• Issued: 11/26/2013
• Est. Priority Date: 09/12/2006
• Status: Active Grant

• Current Assignee: NVIDIA Corporation
• Original Assignee: NVIDIA Corporation
• Inventors: King, Gary C., Hutchins, Edward A., Toksvig, Michael J. M.
• Primary Examiner(s): Alavi, Amir
• Application Number:
  US11/520,144
• Time in Patent Office: 2,632 Days
• Field of Search: 382162-167, 382/232, 382/233, 382/253, 382/298, 345/430, 345/431, 345/503, 345/555, 345/582, 345/583, 345/587, 345/606, 345/608, 358/515, 358/518, 358/523, 358/525, 358/530
• US Class Current: 382/232
• CPC Class Codes: G06T 9/00 Image coding bandwidth or redundancy reduction for static pictures H04N1/41; coding or decoding of static colour picture signals H04N1/64; methods or arrangements for coding, decoding, compressing or decompressing digital video signals H04N19/00
• CPC Class Codes: H04N 1/646 Transmitting or storing colour television type signals, e.g. PAL, Lab; Their conversion into additive or subtractive colour signals or vice versa therefor H04N1/642, H04N1/644 take precedence
• CPC Class Codes: H04N 19/124 Quantisation
• CPC Class Codes: H04N 19/136 Incoming video signal characteristics or properties
• CPC Class Codes: H04N 19/186 the unit being a colour or a chrominance component

Abstract

Image-based data, such as a block of texel data, is accessed. The data includes sets of color component values. A luminance value is computed for each set of color components values, generating a range of luminance values. A first set and a second set of color component values that correspond to the minimum and maximum luminance values are selected from the sets of color component values. A third set of color component values can be mapped to an index that identifies how the color component values of the third set can be decoded using the color component values of the first and second sets. The index value is selected by determining where the luminance value for the third set lies in the range of luminance values.

Claims

1. A method of compressing texel data, said method comprising:

accessing data comprising sets of color component values for a plurality of texels, each of said sets describing a color;

selecting a first set of color component values and a second set of color component values from said sets, wherein said first and second sets define endpoints of a range of colors; and

encoding a third set of color component values by associating said third set with an index value, wherein said third set is decodable using said first and second sets and said index value.

2. The method of claim 1 wherein each of said color component values is represented using a number of bits and further comprising reducing said number of bits prior to said selecting.

3. The method of claim 1 wherein said selecting comprises determining a range of luminance values using said sets of color component values, wherein said first set corresponds to a first endpoint of said range of luminance values and said second set corresponds to a second endpoint of said range of luminance values.

4. The method of claim 3 wherein said encoding comprises determining a luminance value for said third set, wherein said luminance value is between said first and second endpoints of said range of luminance values and wherein said index value is selected according to where said luminance value lies in said range of luminance values.

5. The method of claim 3 wherein said plurality of luminance values comprises a luminance value for each of said sets of color component values and further comprising determining a luminance value for a set of color component values by computing a dot product of color component values in said set.

6. The method of claim 5 further comprising computing said dot product as 0.299R+0.587G+0.114B where R, G and B are red, green and blue component values in said set.

7. The method of claim 5 further comprising computing said dot product by:

shifting a red value in said set one bit left to produce a shifted red value;

shifting a green value in said set two bits left to produce a shifted green value; and

adding said shifted red value and said shifted green value to a blue value in said set.

8. A method of compressing image-based data, said method comprising:

accessing data comprising a plurality of sets of color component values;

determining a plurality of luminance values using said plurality of sets of color component values;

identifying a minimum and a maximum of said luminance values, wherein said minimum and maximum define endpoints of a range of luminance values, and wherein one endpoint of said range corresponds to a first set of color component values and the other endpoint of said range corresponds to a second set of color component values and; and

encoding a third set of color component values in said plurality by associating an index value with said third set, wherein said third set is decodable using said first and second sets and said index value.

9. The method of claim 8 wherein each of said color component values is represented using a number of bits and further comprising reducing said number of bits prior to said selecting.

10. The method of claim 8 wherein said plurality of luminance values comprises a luminance value for each of said sets of color component values and further comprising computing a dot product of color component values in a set of color component values to determine a luminance value for said set.

11. The method of claim 10 further comprising computing said dot product as 0.299R+0.587G+0.1146 where R, G and B are red, green and blue component values in said set.

12. The method of claim 10 further comprising computing said dot product by:

shifting a red value in said set one bit left to produce a shifted red value;

shifting a green value in said set two bits left to produce a shifted green value; and

adding said shifted red and green values in said set to a blue value in said set.

13. The method of claim 8 wherein said first and second sets of color component values are selected from said plurality of sets of color component values.

14. A method of processing image-based data, said method comprising:

accessing data comprising a plurality of sets of color component values;

determining a plurality of luminance values using said plurality of sets of color component values;

identifying first and second endpoints of a range of said plurality of luminance values, wherein said first endpoint corresponds to a minimum of said luminance values that is determined using a first set of color component values and wherein said second endpoint corresponds to a maximum of said luminance values that is determined using a second set of color component values; and

associating an index value with a third set of color component values, wherein said index value is selected by determining where a luminance value for said third set lies in said range of luminance values.

15. The method of claim 14 wherein each of said color component values is represented using a number of bits and further comprising reducing said number of bits prior to said selecting.

16. The method of claim 14 wherein said plurality of luminance values comprises a luminance value for each of said sets of color component values and further comprising determining a luminance value for a set of color component values by computing a dot product of color component values in said set.

17. The method of claim 16 wherein said dot product is computed for said each set of color component values as 0.299R+0.587G+0.1146 where R, G and B are red, green and blue component values in said set.

18. The method of claim 16 further comprising computing said dot product by:

shifting a red value in said set one bit left to produce a shifted red value;

shifting a green value in said set two bits left to produce a shifted green value; and

adding said shifted red and green values in said set to a blue value in said set.

19. The method of claim 14 further comprising:

identifying thresholds within said range, wherein said thresholds demarcate sub-ranges of luminance values within said range and wherein each of said sub-ranges is associated with a respective index value; and

comparing said luminance value for said third set to said thresholds to identify a sub-range that includes said luminance value.

20. The method of claim 14 wherein said first and second sets are selected from said plurality of sets of color component values, wherein said third set is decodable using said first and second sets and said index value.