IP Library › Granted Patent US 11,438,635
Granted Patent B2
US 11,438,635 · App. 17/110,052 · Granted Sep 6, 2022

Flexible tree partitioning processes for visual media coding

Inventors: Li Zhang (San Diego, CA); Kai Zhang (San Diego, CA); Hongbin Liu (Beijing, CN); Hsiao Chiang Chuang (San Diego, CA); Yue Wang (Beijing, CN)
Assignees: BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD.; BYTEDANCE INC.
H04N19/96H04N19/119H04N19/176H04N19/184H04N19/189H04N19/1883H04N19/46
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,438,635
App. No.
17/110,052
Filed
Dec 2, 2020
Granted
Sep 6, 2022
Kind
B2
Examiner
BECK, LERON
Art Unit
2487
USPC
375/240.02
Abstract

Methods, systems and devices for using flexible and efficient partitioning techniques are described. An exemplary method for visual media decoding includes applying, to a current visual media block, a partitioning process that splits the current visual media block into more than four sub-blocks; decoding, based on a bitstream representation, the more than four sub-blocks; and decoding, based on the more than four sub-blocks and the partitioning process, the current visual media block. Another exemplary method for visual media encoding includes receiving input data associated with a current visual media block; applying, to the input data, a partitioning process that splits the current visual media block into more than four sub-blocks; encoding, based on the partitioning process, the more than four sub-blocks; and generating, based on the encoding, a bitstream representation of the current visual media block.

Claims (40)

1. A method of visual media processing, comprising:

applying, during a conversion between a current visual media block and a bitstream of the current visual media block, a one-step partitioning process that splits the current visual media block into more than four non-overlapping sub-blocks at a first depth, wherein a first sub-block of the more than four sub-blocks has a size different from a second sub-block of the more than four sub-blocks;

splitting at least one of the more than four sub-blocks using at least one of a binary tree (BT) partitioning process, a quadtree (QT) partitioning process, a ternary tree (TT) partitioning process, or an extended quadtree (EQT) partitioning process to generate a final partition structure; and

performing the conversion based on the final partition structure,

wherein the bitstream includes one or more indications of a partitioning process that results in the final partition structure from the current visual media block.

2. The method of claim 1 , further comprising:

recursively applying the partitioning process to at least one of the more than four sub-blocks to generate one or more leaf nodes.

3. The method of claim 1 , wherein the recursively applying the partitioning process is implicitly terminated when (i) a size of a leaf node reaches a minimally allowed leaf node size for the partitioning process, or (ii) a depth associated with the at least one of the more than four sub-blocks reaches a maximally allowed depth of the partitioning process.

4. The method of claim 1 , further comprising:

deciding, based on a condition, regarding a selective inclusion of one or more signaling bits for the partitioning process in the bitstream.

5. The method of claim 4 , wherein the partitioning process is used to replace an existing partitioning process that includes a binary tree (BT) partitioning process, a ternary tree (TT) partitioning process or a quadtree (QT) partitioning process.

6. The method of claim 5 , wherein the condition is based on a picture type or a slice type of a picture or a slice comprising the current visual media block, respectively.

7. The method of claim 5 , wherein the condition is based on a size of the current visual media block.

8. The method of claim 5 , wherein the condition is based on a coding mode of the current visual media block.

9. The method of claim 5 , wherein the condition is based on a location of the current visual media block, wherein the location of the current visual media block is at a boundary of a picture, a slice or a tile comprising the current visual media block.

10. The method of claim 5 , wherein the partitioning process is treated in the same way as a quadtree (QT) partitioning process.

11. The method of claim 10 , wherein the partitioning process is treated as an additional partition pattern different from the QT partitioning.

12. The method of claim 5 , wherein the condition comprises an unavailability of a transform matrix for a selected partition size of the partitioning process.

13. The method of claim 1 , wherein the conversion includes encoding the current visual media block into the bitstream.

14. The method of claim 1 , wherein the conversion includes decoding the current visual media block from the bitstream.

15. An apparatus in a video system comprising a processor and a non- transitory memory with instructions thereon, wherein the instructions upon execution by the processor, cause the processor to:

apply, during a conversion between a current visual media block and a bitstream of the current visual media block, a one-step partitioning process that splits the current visual media block into more than four non-overlapping sub-blocks at a first depth, wherein a first sub-block of the more than four sub-blocks has a size different from a second sub-block of the more than four sub-blocks;

split at least one of the more than four sub-blocks using at least one of a binary tree (BT) partitioning process, a quadtree (QT) partitioning process, a ternary tree (TT) partitioning process, or an extended quadtree (EQT) partitioning process to generate a final partition structure; and

perform the conversion based on the final partition structure,

wherein the bitstream includes one or more indications of a partitioning process that results in the final partition structure from the current visual media block.

16. A non-transitory computer-readable storage medium storing instructions that cause a processor to:

apply, during a conversion between a current visual media block and a bitstream of the current visual media block, a one-step partitioning process that splits the current visual media block into more than four non-overlapping sub-blocks at a first depth, wherein a first sub-block of the more than four sub-blocks has a size different from a second sub-block of the more than four sub-blocks;

split at least one of the more than four sub-blocks using at least one of a binary tree (BT) partitioning process, a quadtree (QT) partitioning process, a ternary tree (TT) partitioning process, or an extended quadtree (EQT) partitioning process to generate a final partition structure; and

perform the conversion based on the final partition structure,

wherein the bitstream includes one or more indications of a partitioning process that results in the final partition structure from the current visual media block.

17. A non-transitory computer-readable recording medium storing a bitstream of a video which is generated by a method performed by a video processing apparatus, wherein the method comprises:

applying, during a conversion between a current visual media block and a bitstream of the current visual media block, a one-step partitioning process that splits the current visual media block into more than four non-overlapping sub-blocks at a first depth, wherein a first sub-block of the more than four sub-blocks has a size different from a second sub-block of the more than four sub-blocks;

splitting at least one of the more than four sub-blocks using at least one of a binary tree (BT) partitioning process, a quadtree (QT) partitioning process, a ternary tree (TT) partitioning process, or an extended quadtree (EQT) partitioning process to generate a final partition structure, wherein only one combination of partitioning processes that result in the final partition structure is valid; and

generating the bitstream from the current visual media block based on the final partition structure,

wherein the bitstream includes one or more indications of a partitioning process that results in the final partition structure from the current visual media block.

18. The apparatus of claim 15 , wherein the instructions upon execution by the processor, further cause the processor to:

recursively apply the partitioning process to at least one of the more than four sub-blocks to generate one or more leaf nodes.

19. The apparatus of claim 15 , wherein the recursively applying the partitioning process is implicitly terminated when (i) a size of a leaf node reaches a minimally allowed leaf node size for the partitioning process, or (ii) a depth associated with the at least one of the more than four sub-blocks reaches a maximally allowed depth of the partitioning process.

20. The non-transitory computer-readable storage medium of claim 16 , wherein the instructions further cause the processor to:

recursively apply the partitioning process to at least one of the more than four sub-blocks to generate one or more leaf nodes.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 2, 2020
From: ZHANG, LI; ZHANG, KAI; CHUANG, HSIAO CHIANG
To: BYTEDANCE INC.
Reel/Frame 054521/0878 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 2, 2020
From: LIU, HONGBIN; WANG, YUE
To: BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD.
Reel/Frame 054521/0897 →
Priority Claims (1)
CN PCT/CN2018/089918 · Jun 5, 2018 · national
Continuity (2)
Continuation PCTIB2019054613 · Jun 4, 2019
Related Publication 20210120243A1 · Apr 22, 2021