公开(公告)号：US20250061538A1

公开(公告)日：2025-02-20

申请号：US18805459

申请日：2024-08-14

Applicant: Electronics and Telecommunications Research Institute

Inventor： Youn Hee KIM ,  JooYoung LEE ,  Se Yoon JEONG ,  Jin Soo CHOI ,  Jung Won KANG

IPC: G06T3/067 ,  G06T3/4046

Abstract: A feature map encoding method according to the present disclosure may include generating a feature map from a multi-level feature group; and performing PCA (Picture Component Analysis) transform for the feature map. Here, generating the feature map comprises reshaping the multi-level feature group; and generating the feature map by merging a plurality of base units generated by the reshape.

公开(公告)号：US20250133221A1

公开(公告)日：2025-04-24

申请号：US18912827

申请日：2024-10-11

Applicant: Electronics and Telecommunications Research Institute ,  UNIVERSITY-INDUSTRY COOPERATION GROUP OF KYUNG HEE UNIVERSITY

Inventor： JooYoung LEE ,  Youn Hee KIM ,  Se Yoon JEONG ,  Jin Soo CHOI ,  Jung Won KANG ,  Hui Yong KIM ,  Hye Won JEONG ,  Seung Hwan JANG ,  Yeong Woong KIM ,  Jang Hyun YU

IPC: H04N19/169 ,  G06V10/77 ,  G06V10/82 ,  H04N19/124 ,  H04N19/13 ,  H04N19/189 ,  H04N19/30 ,  H04N19/70

Abstract: A device for decoding a feature map according to the present disclosure comprises an image decoding unit to decode an image from a bitstream; an inverse format conversion unit to restore a feature map latent representation by converting a formation of a decoded image; and a feature map restoration unit to restore a multi-layer feature map from the feature map latent representation. Here, the feature map restoration unit restores the multi-layer feature map based on a learned neural network parameter.

公开(公告)号：US20250054118A1

公开(公告)日：2025-02-13

申请号：US18797916

申请日：2024-08-08

Applicant: Electronics and Telecommunications Research Institute

Inventor： Youn Hee KIM ,  JooYoung LEE ,  Se Yoon JEONG ,  Jin Soo CHOI ,  Jun Won KANG

IPC: G06T5/80 ,  G06T5/60

Abstract: A feature map decoding method according to the present disclosure may include decoding a metadata and the feature map; performing inverse conversion on a decoded feature map; and restoring features from inverse converted features. Here, a feature distortion compensation is performed on at least one of the decoded feature map, the inverse converted features or restored features.