Method and apparatus for filtering video data using a programmable graphics processor

• US 7,508,448 B1
• Filed: 05/29/2003
• Issued: 03/24/2009
• Est. Priority Date: 05/29/2003
• Status: Active Grant

• Current Assignee: NVIDIA Corporation
• Original Assignee: NVIDIA Corporation
• Inventors: Amann, Garry W., Azar, Hassane S., Lew, Stephen D.
• Primary Examiner(s): Tran; Trang U
• Application Number:
  US10/448,590
• Time in Patent Office: 2,126 Days
• Field of Search: 348/448, 348/441, 348449-452, 348/458, 348699-701, 348/604, 348/558, 345/531, 345/501, 345430-431
• US Class Current: 348/448
• CPC Class Codes: H04N 19/86 involving reduction of coding artifacts, e.g. of blockiness
• CPC Class Codes: H04N 7/01 Conversion of standards , e.g. involving analogue television standards or digital television standards processed at pixel level
• CPC Class Codes: H04N 7/012 Conversion between an interlaced and a progressive signal for graphics images G09G2310/0229

Abstract

Video filtering using a programmable graphics processor is described. The programmable graphics processor may be programmed to complete a plurality of video filtering operations in a single pass through a fragment-processing pipeline within the programmable graphics processor. Video filtering functions such as deinterlacing, chroma up-sampling, scaling, and deblocking may be performed by the fragment-processing pipeline. The fragment-processing pipeline may be programmed to perform motion adaptive deinterlacing, wherein a spatially variant filter determines, on a pixel basis, whether a “bob”, a “blend”, or a “weave” operation should be used to process an interlaced image.

Claims

1. A method for performing video filtering, said method comprising:

receiving an input video signal having a plurality of odd and even frames into a shading pipeline within a programmable graphics processor; and

applying a spatially variant deinterlacing function on the input video signal in a single pass of the shading pipeline to produce deinterlaced video frames, wherein the spatially variant deinterlacing function is applied, based on characteristics of video image data for a current interlaced frame, to video image data associated with a corresponding pixel within a previous interlaced frame.

2. The method of claim 1, wherein the spatially variant deinterlacing function applies a plurality of bob operations, a plurality of blend operations, and a plurality of weave operations.

3. The method of claim 2, wherein a decision for determining whether to apply a bob operation, a blend operation, or a weave operation to a pixel is based on detected motion relative to the pixel.

4. The method of claim 1, wherein the video filtering method further comprises:

applying at least one of a scaling function, a chroma up-sampling function, a deblocking function, and a color space conversion function within the shading pipeline.

5. The method of claim 4, wherein the scaling function obtains fractional portions from texture coordinates associated with a pixel and uses the fractional portions as a filter kernel phase to access a plurality of filter values.

6. The method of claim 4, wherein a filter kernel phase used to access a plurality of filter values is computed using the shading pipeline.

7. The method of claim 4, wherein the deblocking function is a spatially dependent function.

8. The method of claim 7, wherein the deblocking function is a data dependent function.

9. The method of claim 1, wherein the video filtering method further comprises:

applying at least one of a scaling function, a chroma up-sampling function, a deblocking function, and a color space conversion function within the shading pipeline within the single pass of the shading pipeline.

10. The method of claim 1, wherein the video filtering method further comprises:

applying at least one of a reverse gamma correction function and a gamma correction function within the shading pipeline.

11. The method of claim 1, wherein the step of applying includes the application of a filter by the shading pipeline, wherein the filter is stored as an array of filter values within a filter texture map in the graphics memory and is associated with a unique texID (texture identification).

12. A method for performing video filtering, the method comprising:

receiving an input video signal having a plurality of odd and even frames into a shading pipeline within a programmable graphics processor; and

applying a spatially variant deinterlacing function on the input video signal in a single pass of the shading pipeline using a shader program to produce deinterlaced video data, wherein the spatially variant deinterlacing function is applied, based on characteristics of video image data for a current interlaced frame, to video image data associated with a corresponding pixel within a previous interlaced frame.

13. The method of claim 12, wherein the spatially variant deinterlacing function applies a plurality of bob operations, a plurality of blend operations, and a plurality of weave operations.

14. The method of claim 12, wherein the shading pipeline determines whether to apply a bob operation, a blend operation, or a weave operation to a pixel within the video input signal based on detected motion relative to the pixel.

15. The method of claim 12, wherein the video filtering method further comprises:

applying at least one of a scaling function, a chroma up-sampling function, a deblocking function, and a color space conversion function within the shading pipeline using the shader program.

16. The method of claim 15, wherein the scaling function obtains a filter kernel phase to access a plurality of filter values.

17. The method of claim 15, wherein a filter kernel phase used to access a plurality of filter values for a pixel is produced using fractional portions extracted from texture coordinates associated with the pixel.

18. The method of claim 15, wherein a filter kernel phase used to access a plurality of filter values is computed using the shading pipeline.

19. The method of claim 15, wherein the deblocking function is a spatially dependent function.

20. The method of claim 19, wherein the deblocking function is a data dependent function.

21. The method of claim 12, wherein the video filtering method further comprises:

applying at least one of a reverse gamma correction function and a gamma correction function within the shading pipeline.