System and method for video encoding using adaptive loop filter

• US 8,326,075 B2
• Filed: 12/05/2008
• Issued: 12/04/2012
• Est. Priority Date: 09/11/2008
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Wilkins, Paul, Bankoski, James, Xu, Yaowu
• Primary Examiner(s): Chu, Randolph I
• Application Number:
  US12/329,070
• Publication Number:
  US 20100061645A1
• Time in Patent Office: 1,460 Days
• Field of Search: None
• US Class Current: 382/268
• CPC Class Codes: H04N 19/117 Filters, e.g. for pre-processing or post-processing sub-band filter banks H04N19/635
• CPC Class Codes: H04N 19/139 Analysis of motion vectors, e.g. their magnitude, direction, variance or reliability
• CPC Class Codes: H04N 19/159 Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/50 using predictive coding H04N19/61 takes precedence
• CPC Class Codes: H04N 19/513 Processing of motion vectors
• CPC Class Codes: H04N 19/65 using error resilience
• 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

Methods for decoding frames of compressed video information are disclosed. Each frame includes a plurality of blocks having a prediction stage parameter and a residual error attribute. One method includes categorizing each of the blocks into categories based on at least one of the prediction stage parameter or the residual error attribute, determining a loop filter strength value to each of the categories, and filtering the boundary adjacent to at least one of the blocks using the loop filter strength assigned to the category in which that block is categorized.

Claims

1. A method for reducing blocking artifacts at the boundary between adjacent blocks reconstructed from a frame of compressed video information that includes a prediction stage parameter with respect to at least one of the blocks, the method comprising:

reconstructing the at least one block based on the prediction stage parameter;

computing a residual error attribute from the reconstructed block;

categorizing the at least one block into one of a plurality of categories based on at least one of the prediction stage parameter or the residual error attribute;

identifying one or more filter modifiers for the category in which the at least one block is categorized based on at least one of the prediction stage parameter or the residual error attribute associated with that category;

computing a filter strength value based on a baseline filter strength and, the identified one or more filter modifiers; and

filtering the boundary adjacent to the at least one block using the selected computed filter strength value.

2. The method of claim 1, wherein the prediction stage parameter includes at least one of a reference frame type and a prediction mode.

3. The method of claim 2, wherein the reference frame type is one of intra-frame, last frame, golden frame and alternate reference frame.

4. The method of claim 2, wherein the prediction mode is one of non-split mode and split mode.

5. The method of claim 1, wherein the residual error attribute is one of a null motion vector and a non-zero motion vector.

6. The method of claim 1 wherein the one or more loop filter modifiers are determined at the frame level.

7. The method of claim 6, further comprising:

determining whether the frame being decoded is a key frame; and

wherein at least some of the loop filter modifiers are set to a default value if the frame being decoded is a key frame.

8. The method of claim 6, further comprising:

determining whether the frame being decoded specifies a loop filter strength value condition, wherein the loop filter strength value condition indicates that a pre-determined loop filter strength value is to be used for the plurality of categories; and

if the frame being decoded specifies a loop filter strength value condition, then assigning the pre-determined loop filter strength value to the plurality of categories.

9. The method of claim 6, further comprising:

determining whether the frame being decoded specifies a single loop filter condition; and

if the frame being decoded specifies a single loop filter condition, then assigning a single loop filter strength value to blocks of the frame without regard to the prediction stage parameter and residual error attribute associated with the at least one block.

10. A method for decoding frames of compressed video information, each frame including a plurality of blocks having a prediction stage parameter and a residual error attribute, the method comprising:

categorizing the blocks of at least one of the frames into categories based on at least one of the prediction stage parameter or the residual error attribute;

determining a loop filter strength value for each of the categories using the steps of;

determining a baseline loop filter strength value specified in the at least one frame,

identifying one or more loop filter modifiers for the given category based on at least one of the prediction stage parameter or the residual error attribute associated with that category, and

determining the loop filter strength value based on the one or more identified loop filter modifiers and the baseline loop filter strength value; and

filtering the boundary adjacent to at least one of the blocks using the loop filter strength value assigned to the category in which that block is categorized.

11. The method of claim 10, wherein determining the loop filter strength value further includes:

adding the one or more loop filter modifiers to the baseline loop filter strength value.

12. The method of claim 10, wherein the prediction stage parameter includes at least one of a reference frame type or a prediction mode.

13. The method of claim 12, wherein the reference frame type is one of intra-frame, last frame, golden frame and alternate reference frame.

14. The method of claim 12, wherein the prediction mode is one of non-split mode and split mode.

15. The method of claim 10, wherein the residual error attribute is one of a null motion vector and a non-zero motion vector.

16. The method of claim 10, wherein the loop filter modifiers are determined based on delta values specified in at least one of the frames.

17. The method of claim 16, further comprising:

determining whether the frame being decoded is a key frame; and

wherein at least some of the loop filter modifiers are set to a default value if the frame being decoded is a key frame.

18. The method of claim 17, further comprising:

determining whether the frame being decoded specifies a loop filter strength value condition for one or more of the categories, wherein the loop filter strength value condition indicates that a pre-determined loop filter strength value is to be used for the one or more categories; and

if the frame being decoded specifies the loop filter strength value condition for one or more of the categories, then assigning the pre-determined loop filter strength value to the one or more categories.

19. The method of claim 10, further comprising:

determining whether one or more of the frames being decoded specify a single loop filter condition; and

if the one or more frames being decoded specify a single loop filter condition, then assigning a single loop filter strength value to blocks of the one or more frames without regard to category.

20. The method of claim 10, wherein identifying the one or more loop filter modifiers includes:

identifying a first loop filter modifier based on the prediction stage parameter; and

identifying a second loop filter modifier based on the residual error attribute.

21. The method of claim 20, wherein determining the loop filter strength value includes adding the baseline loop filter strength value, the first loop filter modifier, and the second loop filter modifier.

22. A method for decoding a video stream including a plurality of frames, at least some frames having a plurality of blocks, the method comprising:

decoding, from the video stream, a baseline filter strength value associated with a current frame of the plurality of frames;

determining a plurality of loop filter modifiers for the current frame based on one or more delta values decoded from the video stream;

decoding a prediction stage parameter and a residual error attribute associated with a block of the current frame;

selecting one or more of the loop filter modifiers based on at least one of the prediction stage parameter or the residual error attribute;

determining a loop filter strength based on the selected loop filter modifiers and the baseline filter strength value; and

applying a filter to the block at the determined loop filter strength.

23. The method of claim 22, wherein determining the plurality of loop filter modifiers for the current frame is further based on a combination of one or more delta values associated with a previous frame and one or more delta values associated with the current frame.

24. The method of claim 22, wherein the plurality of loop filter modifiers includes at least one of filter type, filter coefficients, or filter taps and applying the filter to the block is based on the filter type, filter coefficients, or filter taps included in the selected loop filter modifiers.