公开(公告)号：US20220417507A1

公开(公告)日：2022-12-29

申请号：US17779093

申请日：2020-11-23

Applicant: Electronics and Telecommunications Research Institute ,  KWANGWOON UNIVERSITY INDUSTRY-ACADEMIC COLLABORATION FOUNDATION

Inventor： Seoung-Jun OH ,  Woong LIM ,  Gun BANG ,  Tae-Jin LEE ,  Dong-Gyu SIM ,  Jun-Taek PARK ,  Min-Hun LEE ,  Jong-Seok LEE

IPC: H04N19/117 ,  H04N19/80 ,  H04N19/105 ,  H04N19/176

Abstract: Disclosed herein are a video processing method, apparatus and storage medium using the resolution of a reference picture. Filtering is performed based on resolutions of reference pictures of blocks forming a boundary. The resolutions of the reference pictures may define filtering. Depending on whether the resolutions of the reference pictures are different from each other, one or more filter selection parameters for filtering may be determined or updated. Further, based on whether the resolutions of the reference pictures are different from each other, the boundary strength of the filter may be determined. Because filtering is defined using the resolutions of reference pictures, performance of encoding and/or decoding may be improved.

公开(公告)号：US20220321890A1

公开(公告)日：2022-10-06

申请号：US17629281

申请日：2020-07-23

Applicant: Electronics and Telecommunications Research Institute ,  INDUSTRY-ACADEMIA COOPERATION GROUP OF SEJONG UNIVERSITY

Inventor： Woong LIM ,  Jin-Young LEE ,  Gun BANG ,  Hui-Yong KIM

IPC: H04N19/137 ,  H04N19/159 ,  H04N19/182 ,  H04N19/176

Abstract: Disclosed herein are a method, an apparatus, and a storage medium for image encoding/decoding using geometric partitioning. A target block may be geometrically partitioned, and multiple partitioned regions may be generated from geometric partitioning. Each partitioned region may have a non-square and a non-rectangular shape. Different prediction schemes may be applied to different partitioned regions. The different prediction schemes for multiple partitioned regions may be signaled as explicit information, and may be derived using information about a coding parameter for the target block or the like.

公开(公告)号：US20210203921A1

公开(公告)日：2021-07-01

申请号：US17134092

申请日：2020-12-24

Applicant: Electronics and Telecommunications Research Institute

Inventor： Gun BANG ,  Woong LIM

IPC: H04N19/105 ,  H04N19/119 ,  H04N19/46 ,  H04N19/176

Abstract: Disclosed herein are a method, apparatus and storage medium for image encoding/decoding using a reference picture. A picture is divided into multiple sub-pictures. The sub-pictures of the picture share a reference picture list. In order to determine sub-pictures that share the reference picture list, specific information is signaled from an encoding apparatus to a decoding apparatus. For each sub-picture that does not share a reference picture list for an additional sub-picture, a reference picture list for the sub-picture is separately configured.

公开(公告)号：US20210136416A1

公开(公告)日：2021-05-06

申请号：US16768054

申请日：2018-11-27

Applicant: Electronics and Telecommunications Research Institute

Inventor： Youn-Hee KIM ,  Hui-Yong KIM ,  Seung-Hyun CHO ,  Jin-Wuk SEOK ,  Joo-Young LEE ,  Woong LIM ,  Jong-Ho KIM ,  Dae-Yeol LEE ,  Se-Yoon JEONG ,  Jin-Soo CHOI

IPC: H04N19/61 ,  H04N19/105 ,  H04N19/176 ,  G06N3/08

Abstract: Disclosed herein are a video decoding method and apparatus and a video encoding method and apparatus. A transformed block is generated by performing a first transformation that uses a prediction block for a target block. A reconstructed block for the target block is generated by performing a second transformation that uses the transformed block. The prediction block may be a block present in a reference image, or a reconstructed block present in a target image. The first transformation and the second transformation may be respectively performed by neural networks. Since each transformation is automatically performed by the corresponding neural network, information required for a transformation may be excluded from a bitstream.

公开(公告)号：US20210084290A1

公开(公告)日：2021-03-18

申请号：US16772443

申请日：2018-12-13

Applicant: Electronics and Telecommunications Research Institute

Inventor： Seung-Hyun CHO ,  Joo-Young LEE ,  Youn-Hee KIM ,  Jin-Wuk SEOK ,  Woong LIM ,  Jong-Ho KIM ,  Dae-Yeol LEE ,  Se-Yoon JEONG ,  Hui-Yong KIM ,  Jin-Soo CHOI

IPC: H04N19/105 ,  H04N19/176 ,  H04N19/159 ,  H04N19/85

Abstract: Disclosed herein are a method and apparatus for video decoding and a method and apparatus for video encoding. A prediction block for a target block is generated by predicting the target block using a prediction network, and a reconstructed block for the target block is generated based on the prediction block and a reconstructed residual block. The prediction network includes an intra-prediction network and an inter-prediction network and uses a spatial reference block and/or a temporal reference block when it performs prediction. For learning in the prediction network, a loss function is defined, and learning in the prediction network is performed based on the loss function.

公开(公告)号：US20190306526A1

公开(公告)日：2019-10-03

申请号：US16374601

申请日：2019-04-03

Applicant: Electronics and Telecommunications Research Institute

Inventor： Seung-Hyun CHO ,  Je-Won KANG ,  Na-Young KIM ,  Jung-Kyung LEE ,  Joo-Young LEE ,  Hyunsuk KO ,  Youn-Hee KIM ,  Jong-Ho KIM ,  Jin-Wuk SEOK ,  Dae-Yeol LEE ,  Woong LIM ,  Se-Yoon JEONG ,  Hui-Yong KIM ,  Jin-Soo CHOI

IPC: H04N19/513 ,  H04N19/105 ,  H04N19/132 ,  G06N20/00

Abstract: Disclosed herein are an inter-prediction method and apparatus using a reference frame generated based on deep learning. In the inter-prediction method and apparatus, a reference frame is selected, and a virtual reference frame is generated based on the selected reference frame. A reference picture list is configured to include the generated virtual reference frame, and inter prediction for a target block is performed based on the virtual reference frame. The virtual reference frame may be generated based on a deep-learning network architecture, and may be generated based on video interpolation and/or video extrapolation that use the selected reference frame.

公开(公告)号：US20250055977A1

公开(公告)日：2025-02-13

申请号：US18922993

申请日：2024-10-22

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Gun BANG ,  Woong LIM ,  Hui-Yong KIM

IPC: H04N19/105 ,  H04N19/174 ,  H04N19/176 ,  H04N19/46

Abstract: Disclosed herein are a method, apparatus, and storage medium for image encoding/decoding using a reference picture. A reference picture list is configured for encoding/decoding of a target block. Upon configuring the reference picture list, a sequence parameter set and a slice header are used. A picture is divided into sub-pictures and/or slices, and reference picture lists are configured for the sub-pictures and/or slices resulting from division. Reference picture lists for slices may be related to each other, and a current reference picture and a future reference picture in one reference picture list may be used in an additional reference picture list.

公开(公告)号：US20240430442A1

公开(公告)日：2024-12-26

申请号：US18821623

申请日：2024-08-30

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Woong LIM ,  Gun BANG

IPC: H04N19/159 ,  H04N19/176 ,  H04N19/186

Abstract: Disclosed herein are a method, an apparatus and a storage medium for image encoding/decoding using a palette mode. Prediction using a palette mode may be performed for a target block. Various types of predictions may be performed for the target block, and pieces of information to be signaled from an encoding apparatus to a decoding apparatus may be changed depending on the type of prediction to be performed for the target block. Determination of the pieces of information to be signaled may be optimized for the case where a palette mode is to be used for the target block, and thus efficiency of image encoding and decoding may be improved.

公开(公告)号：US20240406436A1

公开(公告)日：2024-12-05

申请号：US18697343

申请日：2022-09-29

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Dong-Hyun KIM ,  Se-Yoon JEONG ,  Hyoung-Jin KWON ,  Youn-Hee KIM ,  Jong-Ho KIM ,  Ji-Hoon DO ,  Joo-Young LEE ,  Woong LIM ,  Jin-Soo CHOI ,  Tae-Jin LEE

IPC: H04N19/52 ,  H04N19/109 ,  H04N19/139 ,  H04N19/176 ,  H04N19/573

Abstract: Disclosed herein are a method, an apparatus and a storage medium for image encoding/decoding. In typical image encoding/decoding methods, a decoder-side motion information derivation method may be limitedly used. Therefore, the improvement of encoding efficiency attributable to the decoder-side motion information derivation method may also be limited. In embodiments, a motion information search method used in inter-prediction is disclosed. With the use of various motion search methods, encoding efficiency in inter-prediction may be improved.

公开(公告)号：US20230297833A1

公开(公告)日：2023-09-21

申请号：US18306075

申请日：2023-04-24

Applicant: Electronics and Telecommunications Research Institute

Inventor： Seung-Hyun CHO ,  Youn-Hee KIM ,  Jin-Wuk SEOK ,  Joo-Young LEE ,  Woong LIM ,  Jong-Ho KIM ,  Dae-Yeol LEE ,  Se-Yoon JEONG ,  Hui-Yong KIM ,  Jin-Soo CHOI ,  Je-Won KANG

IPC: G06N3/08 ,  G06N3/04 ,  G06V10/82 ,  G06F18/214

CPC classification number: G06N3/08 ,  G06F18/214 ,  G06N3/04 ,  G06V10/82

Abstract: Disclosed herein are a method and apparatus for compressing learning parameters for training of a deep-learning model and transmitting the compressed parameters in a distributed processing environment. Multiple electronic devices in the distributed processing system perform training of a neural network. By performing training, parameters are updated. The electronic device may share the updated parameter thereof with additional electronic devices. In order to efficiently share the parameter, the residual of the parameter is provided to the additional electronic devices. When the residual of the parameter is provided, the additional electronic devices update the parameter using the residual of the parameter.