SYSTEM AND METHOD FOR IN-LOOP DEBLOCKING IN SCALABLE VIDEO CODING

• US 20080159404A1
• Filed: 12/28/2007
• Published: 07/03/2008
• Est. Priority Date: 12/28/2006
• Status: Active Grant

• Current Assignee: Vidyo Incorporated (Enghouse Systems Limited)
• Original Assignee: Vidyo Incorporated (Enghouse Systems Limited)
• Inventors: Eleftheriadis, Alexandros, Shapiro, Ofer, Hong, Danny
• Granted Patent: US 8,577,168 B2
• Time in Patent Office: Days
• US Class Current: 375/240.23
• CPC Class Codes: H04N 19/30 using hierarchical techniques, e.g. scalability H04N19/63 takes precedence
• CPC Class Codes: H04N 19/50 using predictive coding H04N19/61 takes precedence
• CPC Class Codes: H04N 19/82 involving filtering within a prediction loop
• CPC Class Codes: H04N 19/86 involving reduction of coding artifacts, e.g. of blockiness

Abstract

A system and a method for deblocking a reconstructed/decoded picture in a scalable video encoding/decoding system is provided. Deblocking is accomplished by applying a filter to smooth pixel values adjacent to a boundary shared by two blocks. The type of the filter applied depends on quantization parameter (QP) values assigned to the two blocks. An enhancement layer (EL) block is assigned a QP value based on its coded information and the QP value of its corresponding base layer (BL) block(s).

Claims

1. A method for deblocking a reconstructed/decoded picture in a scalable video encoding/decoding system in which deblocking is accomplished by applying a filter to smooth pixel values adjacent to a boundary of a block Block_C shared with another block Block_X, wherein the number of pixels that are filtered and the type of filtering depend at least on quantization parameter (QP) values QP_X and QP_C assigned to blocks Block_C and Block_X, respectively, the method comprising:

(1a) if neither Block_X nor Block_C has any non-zero coefficient, and if Block_X and Block_C do not have motion vectors that are different by more than a half pixel, then

(2) if the residue of Block_X is predicted only from one or more corresponding base layer (BL) blocks then setting QP_X equal to a weighted average of a QP derived from the one or more corresponding BL blocks and the QP of Block_X, otherwise setting QP_X equal to the QP of Block_X, and

(3) if the residue of Block_C is predicted only from one or more corresponding BL blocks then setting QP_C equal to a weighted average of a QP derived from the said one or more corresponding BL blocks and the QP of Block_C, otherwise setting QP_C equal to the QP of Block_C,

(1b) otherwise, setting QP_X equal to the QP of Block_X and setting QP_C equal to the QP of Block_C.

2. The method of claim 1 wherein the QP derived from the one or more corresponding BL blocks for Block_X or Block_C is a weighted average of the QPs of the BL blocks.

3. The method of claim 1 wherein the QP derived from the one or more corresponding BL blocks for Block_X or Block_C is the minimum of the QPs of the BL blocks.

4. The method of claim 1 wherein the QP derived from the one or more corresponding BL blocks for Block_X or Block_C is the QP of the BL block that has the largest overlap with Block_X or Block_C, respectively.

5. The method of claim 2 wherein the BL blocks themselves are predicted from another lower base layer (BL′

), and wherein their QP is derived as in claim 1 with the current layer replaced by BL and the base layer replaced by BL′

.

6. The method of claim 1 wherein the weighting function is signaled in the bitstream using appropriate flags or parameters.

7. The method of claim 2 wherein the weighting function is signaled in the bitstream using appropriate flags or parameters.

8. Computer readable media comprising a set of instructions to perform the steps recited in at least one of the method claims 1-7.

13. The device of claim 9 wherein the BL blocks themselves are predicted from another lower base layer (BL′

), and wherein their QP is derived as in claim 1 with the current layer replaced by BL and the base layer replaced by BL′

.

9. A device for deblocking a reconstructed/decoded picture in a scalable video encoding/decoding system, the device comprising:

a filter which smoothes pixel values adjacent to a boundary of a block Block_C shared with another block Block_X, wherein the number of pixels that are filtered and the type of filtering depend at least on quantization parameter (QP) values QP_X and QP_C assigned to blocks Block_C and Block_X, respectively, and wherein

(1a) if neither Block_X nor Block_C has any non-zero coefficient, and if Block_X and Block_C do not have motion vectors that are different by more than a half pixel, then

(2) if the residue of Block_X is predicted only from one or more corresponding base layer (BL) blocks then QP_X is set equal to a weighted average of a QP derived from the one or more corresponding BL blocks and the QP of Block_X, otherwise QP_X is set equal to the QP of Block_X, and

(3) if the residue of Block_C is predicted only from one or more corresponding BL blocks then QP_C is set equal to a weighted average of a QP derived from the said one or more corresponding BL blocks and the QP of Block_C, otherwise QP_C is set equal to the QP of Block_C,

(1b) otherwise, QP_X is set equal to the QP of Block_X and QP_C is set equal to the QP of Block_C.

10. The device of claim 9 wherein the QP derived from the one or more corresponding BL blocks for Block_X or Block_C is a weighted average of the QPs of the BL blocks.

11. The device of claim 9 wherein the QP derived from the one or more corresponding BL blocks for Block_X or Block_C is the minimum of the QPs of the BL blocks.

12. The device of claim 9 wherein the QP derived from the one or more corresponding BL blocks for Block_X or Block_C is the QP of the BL block that has the largest overlap with Block_X or Block_C, respectively.

14. The device of claim 9 wherein the weighting function is signaled in the bitstream using appropriate flags or parameters.

15. The device of claim 10 wherein the weighting function is signaled in the bitstream using appropriate flags or parameters.