IP Library › Granted Patent US 12,549,765
Granted Patent B2
US 12,549,765 · App. 17/616,092 · Granted Feb 10, 2026

Image encoding/decoding method and apparatus

Inventors: Jung Won Kang (Daejeon, KR); Ha Hyun Lee (Seoul, KR); Sung Chang Lim (Daejeon, KR); Jin Ho Lee (Daejeon, KR); Hui Yong Kim (Daejeon, KR); Byeung Woo Jeon (Seongnam-si, KR); Jee Yoon Park (Seoul, KR)
Assignees: Electronics and Telecommunications Research Institute; RESEARCH & BUSINESS FOUNDATION SUNGKYUNKWAN UNIVERSITY
H04N19/60H04N19/119H04N19/176H04N19/91H04N19/124H04N19/18
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 12,549,765
App. No.
17/616,092
Granted
Feb 10, 2026
Kind
B2
Abstract

Disclosed herein is an image encoding/decoding method and apparatus. The image decoding method includes obtaining subblock based transform (SBT) use information, obtaining at least one of SBT division information, SBT division direction information or SBT position information, when the SBT use information indicates use of an SBT, and performing the SBT with respect to a current block based on at least one of the SBT division information, the SBT division direction information or the SBT position information. The SBT use information may be obtained when a width of the current block and a height of the current block are less than a maximum transform size.

Claims (55)

1 . An image decoding method comprising:

obtaining subblock based transform (SBT) use information;

obtaining at least one of SBT division information, SBT division direction information or SBT position information, when the SBT use information indicates use of an SBT; and

performing the SBT with respect to a current block based on at least one of the SBT division information, the SBT division direction information or the SBT position information,

wherein the SBT use information is obtained when a width of the current block and a height of the current block are less than a maximum transform size,

wherein the obtaining of the SBT use information comprises:

deriving probability model information of the SBT use information; and

performing entropy-decoding based on the probability model information to obtain the SBT use information,

wherein the deriving of the probability model information of the SBT use information comprises:

setting the probability model information of the SBT use information to a first value when a product of a width and a height of the current block is less than 256;

setting the probability model information of the SBT use information to the first value when the product of the width and the height of the current block is equal to 256; and

setting the probability model information of the SBT use information to a second value when the product of the width and the height of the current block is greater than 256,

wherein the SBT division information indicates a method of dividing the current block for the SBT, the SBT division direction information indicates a division direction for the SBT, and the SBT position information indicates which of subblocks divided from the current block is transformed based on the SBT division information and the SBT division direction information, and

wherein the obtaining of the SBT division direction information comprises:

deriving probability model information of the SBT division direction information based on a shape of the current block when the width and the height of the current block are not equal; and

performing entropy-decoding on a bitstream using the probability model information of the SBT division direction information to obtain the SBT division direction information.

2 . The image decoding method of claim 1 , wherein the obtaining of at least one of the SBT division information, the SBT division direction information or the SBT position information comprises deriving the SBT division direction information based on the shape of the current block.

3 . The image decoding method of claim 2 , wherein when the current block is a rectangular block, the SBT division direction information is determined as a first value or a second value depending on a direction of a long side of the rectangular block.

4 . The image decoding method of claim 1 , further comprising: determining the SBT use information indicates no use of the SBT when the current block is a chroma block.

5 . An image encoding method comprising:

determining whether a subblock based transform (SBT) is used for a current block;

encoding SBT use information based on the determination; and

encoding at least one of SBT division information, SBT division direction information or SBT position information, when the SBT use information indicates use of the SBT,

wherein the SBT use information is encoded when a width of the current block and a height of the current block are less than a maximum transform size,

wherein the encoding of the SBT use information comprises:

determining probability model information of the SBT use information; and

performing entropy-encoding based on the probability model information to encode the SBT use information, and

wherein the determining of the probability model information of the SBT use information comprises:

setting the probability model information of the SBT use information to a first value when a product of a width and a height of the current block is less than 256;

setting the probability model information of the SBT use information to the first value when the product of the width and the height of the current block is equal to 256; and

setting the probability model information of the SBT use information to a second value when the product of the width and the height of the current block is greater than 256,

wherein the SBT division information indicates a method of dividing the current block for the SBT, the SBT division direction information indicates a division direction for the SBT, and the SBT position information indicates which of subblocks divided from the current block is transformed based on the SBT division information and the SBT division direction information, and

wherein the encoding of the SBT division direction information comprises:

deriving probability model information of the SBT division direction information based on a shape of the current block when the width and the height of the current block are not equal; and

performing entropy-encoding using the probability model information of the SBT division direction information to encode the SBT division direction information.

6 . A method of transmitting data comprising a bitstream for an image, the method comprising:

obtaining the bitstream for the image; and

transmitting the data of the bitstream,

wherein the bitstream is generated by performing steps of:

determining whether a subblock based transform (SBT) is used for a current block;

encoding SBT use information based on the determination; and

encoding at least one of SBT division information, SBT division direction information or SBT position information, when the SBT use information indicates use of the SBT,

wherein the SBT use information is encoded when a width of the current block and a height of the current block are less than a maximum transform size,

wherein the encoding of the SBT use information comprises:

determining probability model information of the SBT use information; and

performing entropy-encoding based on the probability model information to encode the SBT use information,

wherein the determining of the probability model information of the SBT use information comprises:

setting the probability model information of the SBT use information to a first value when a product of a width and a height of the current block is less than 256;

setting the probability model information of the SBT use information to the first value when the product of the width and the height of the current block is equal to 256; and

setting the probability model information of the SBT use information to a second value when the product of the width and the height of the current block is greater than 256,

wherein the SBT division information indicates a method of dividing the current block for the SBT, the SBT division direction information indicates a division direction for the SBT, and the SBT position information indicates which of subblocks divided from the current block is transformed based on the SBT division information and the SBT division direction information, and

wherein the encoding of the SBT division direction information comprises:

deriving probability model information of the SBT division direction information based on a shape of the current block when the width and the height of the current block are not equal; and

performing entropy-encoding using the probability model information of the SBT division direction information to encode the SBT division direction information.

7 . The image encoding method of claim 5 , wherein the encoding of at least one of the SBT division information, the SBT division direction information or the SBT position information comprises determining whether to skip encoding the SBT division direction information based on the shape of the current block.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 2, 2021
From: KANG, JUNG WON; LEE, HA HYUN; LIM, SUNG CHANG; LEE, JIN HO; KIM, HUI YONG; JEON, BYEUNG WOO; PARK, JEE YOON
To: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE; RESEARCH & BUSINESS FOUNDATION SUNGKYUNKWAN UNIVERSITY
Reel/Frame 058273/0699 →
Priority Claims (1)
KR 10-2019-0075981 · Jun 25, 2019 · national
Continuity (1)
Related Publication 20220321915A1 · Oct 6, 2022
References Cited (40)
US 20160330467A1 · Park · 2016 [cited by examiner]
US 20190141319A1 · Moon · 2019 [cited by examiner]
US 20200260070A1 · Yoo · 2020 [cited by examiner]
US 20200275130A1 · Bokov · 2020 [cited by examiner]
US 20210235072A1 · Ko · 2021 [cited by examiner]
US 20220046247A1 · Yoo · 2022 [cited by examiner]
US 20220070459A1 · Zhang · 2022 [cited by examiner]
US 20220078428A1 · Le Leannec · 2022 [cited by examiner]
US 20220086490A1 · Koo · 2022 [cited by examiner]
US 20220109875A1 · Koo · 2022 [cited by examiner]
US 20220109877A1 · Choi · 2022 [cited by examiner]
US 20220109878A1 · Koo · 2022 [cited by examiner]
US 20220109886A1 · Xiu · 2022 [cited by examiner]
US 20220232221A1 · Lee · 2022 [cited by examiner]
US 20220248017A1 · Lim · 2022 [cited by examiner]
US 20220295061A1 · Zhang · 2022 [cited by examiner]
US 20220377337A1 · Zhang · 2022 [cited by examiner]
US 20220417529A1 · Zhang · 2022 [cited by examiner]
US 20230224462A1 · Zhao · 2023 [cited by examiner]
US 20230403401A1 · Lee · 2023 [cited by examiner]
CN 105765990A · 2016 [cited by applicant]
CN 109218729A · 2019 [cited by applicant]
GB 2582929A · 2020 [cited by applicant]
KR 100772576B1 · 2007 [cited by applicant]
KR 1020190033036A · 2019 [cited by applicant]
WO WO2020076143A1 · 2020 [cited by examiner]
WO WO2020117902A1 · 2020 [cited by examiner]
WO WO2020145381A1 · 2020 [cited by examiner]
WO WO2020166480A1 · 2020 [cited by examiner]
WO WO2020180769A1 · 2020 [cited by examiner]
WO WO2020185036A1 · 2020 [cited by examiner]
WO WO2020185468A1 · 2020 [cited by examiner]
WO WO2020190835A1 · 2020 [cited by examiner]
WO WO2020197957A1 · 2020 [cited by examiner]
Bross, Benjamin et al. Versatile Video Coding (Draft 5). Joint Video Experts Team (JVET) of ITU-T SG 16 WP 3 and ISO/IEC JTC 1/SC 29/WG 11. JVET-N1001-v10. 14th Meeting: Geneva, CH. pp. 1-407, Mar. 19, 2019 (Year: 2019). [cited by examiner]
{Bross, Benjamin et al. Versatile Video Coding (Draft 5). Joint Video Experts Team (JVET) of ITU-T SG 16 WP 3 and ISO/IEC JTC 1/SC 29/WG 11. JVET-N1001-Version 8. 14th Meeting: Geneva, CH. pp. 1-407, meeting held Mar. 1… [cited by examiner]
Benjamin Bross et al., Versatile Video Coding (Draft 5), JVET-N1001-v8, Joint Video Experts Team (JVET) of ITU- T SG 16 WP 3 and ISO/IEC JTC 1/SC 29/WG 11, 14th Meeting: Geneva, CH, Mar. 19-27, 2019. [cited by applicant]
Jason Jung et al., Non-CE6: Modifications on sub-block transform, JVET-N0190-v2, Joint Video Experts Team (JVET) of ITU-T SG 16 WP 3 and ISO/IEC JTC 1/SC 29/WG 11, 14th Meeting: Geneva, CH, Mar. 19-27, 2019, Wilus Inc. [cited by applicant]
Jianle Chen et al., Algorithm description for Versatile Video Coding and Test Model 5 (VTM 5), JVET-N1002-v2, Joint Video Experts Team (JVET) of ITU-T SG 16 WP 3 and ISO/IEC JTC 1/SC 29/WG 11, 14th Meeting: Geneva, CH, … [cited by applicant]
Yin Zhao et al., CE6: Sub-block transform for inter blocks (CE6.4.1), JVET-M0140-v3, Joint Video Experts Team (JVET) of ITU-T SG 16 WP 3 and ISO/IEC JTC 1/SC 29/WG 11, 13th Meeting: Marrakech, MA, Jan. 9-18, 2019, Huawe… [cited by applicant]