公开(公告)号：US20250154557A1

公开(公告)日：2025-05-15

申请号：US18932174

申请日：2024-10-30

Applicant: GWANGJU INSTITUTE OF SCIENCE AND TECHNOLOGY ,  Seoul National University R&DB Foundation

Inventor： Yeongjae Choi ,  Sunghoon Kwon ,  Yoon Hae Koh ,  Woo Jin Kim ,  Mingweon Chon ,  Hansol Choi

IPC: C12Q1/6806 ,  C12Q1/6869

Abstract: The polymerase chain reaction (PCR) has limitations, including a lack of efficient search strategies and inefficiencies in securing primer sequence lengths, despite its high sensitivity and specificity. Therefore, the present invention has been devised to address these issues, concerning a composition for the selection of the desired nucleic acid and a method for nucleic acid selection using it. By using the composition and method of the present invention, it becomes possible to hierarchically, efficiently, and selectively detect and amplify subsets of oligonucleotides with high specificity, which is expected to be widely utilized in the overall bio/medical field.

公开(公告)号：US12300251B2

公开(公告)日：2025-05-13

申请号：US18070499

申请日：2022-11-29

Applicant: Gwangju Institute of Science and Technology

Inventor： Dong Keon Park ,  Hong Kook Kim ,  Ye Chan Yu

IPC: G10L17/18 ,  G10L21/0272 ,  G10L21/0308 ,  G10L25/18

Abstract: The present invention relates to a speaker diarization technology, and more specifically to, end-to-end speaker diarization system and method through transformer learning having an auxiliary loss-based residual connection to separate speakers by dividing the speakers for time interval, wherein the end-to-end speaker diarization system and method using an auxiliary loss can differentiate and separate speakers through speaker labeling based on the transformer learning using an auxiliary loss even if speaker speeches overlap in a multi-speaker environment.

公开(公告)号：US12291790B2

公开(公告)日：2025-05-06

申请号：US17361838

申请日：2021-06-29

Applicant: GWANGJU INSTITUTE OF SCIENCE AND TECHNOLOGY

Inventor： Chanho Pak ,  Chaekyung Baik ,  Seungwoo Lee

IPC: C25B11/081 ,  B01J23/46 ,  B01J37/04 ,  B01J37/08 ,  B01J37/12 ,  B01J37/16 ,  C25B1/04

Abstract: An embodiment provides a catalyst for water electrolysis which includes an iridium mixed phase formed by physical mixing of two or more selected from metal iridium (Ir), iridium(III) oxide (Ir2O3), or iridium(IV) oxide (IrO2) and has a structure in which nanosheets composed of the iridium mixed phase are stacked. The catalyst for water electrolysis may exhibit high activity and stability for the oxygen evolution reaction in water electrolysis.

公开(公告)号：US12252694B2

公开(公告)日：2025-03-18

申请号：US17377590

申请日：2021-07-16

Applicant: GWANGJU INSTITUTE OF SCIENCE AND TECHNOLOGY

Inventor： In Seop Chang ,  Ji Yeon Kim ,  Ji Yeong Jeong ,  So Young Oh ,  Jin Sung Jeon

IPC: C12N15/74 ,  C12N9/22 ,  C12N15/11

Abstract: Proposed is a genomic editing vector for Eubacterium callanderi for the CRISPR/Cas9 system. The genomic editing vector for Eubacterium callanderi includes a DNA sequence encoding a guide RNA (gRNA) of a cleavage target gene; a DNA sequence encoding a Cas9 protein; a Prbo promoter that is operably linked to the DNA sequence encoding a Cas9 protein; a replication starting point derived from E. coli; a replication starting point for Eubacterium callanderi; and a marker for selecting transgenic strains. It is possible to provide a genomic editing vector for Eubacterium callanderi that can be applied to Eubacterium callanderi strains that are acetogen that is suitable for syngas biorefinery, a method for editing a genome of Eubacterium callanderi strains using the same, and transgenic Eubacterium callanderi strains using the same.

公开(公告)号：US12198198B2

公开(公告)日：2025-01-14

申请号：US17839191

申请日：2022-06-13

Applicant: GIST(Gwangju Institute of Science and Technology)

Inventor： Yong-Gu Lee ,  Sungjae Lee ,  Hojeong Shin ,  Jinsu Kim

IPC: G06V20/56 ,  G06Q40/08 ,  G06V20/70

Abstract: The present disclosure relates to a method and apparatus for split liability assessment of a car accident video using deep learning. The method for split liability assessment of a car accident video using deep learning according to an embodiment of the present disclosure may include: (a) obtaining car accident video data; and (b) outputting accidental split liability information by applying the obtained car accident video data to an accidental split liability determination model.

公开(公告)号：US20240414928A1

公开(公告)日：2024-12-12

申请号：US18437231

申请日：2024-02-08

Applicant: Gwangju Institute Of Science And Technology

Inventor： Kwang Hee LEE ,  Ju Hyeon KIM

IPC: H10K30/50 ,  H10K30/86 ,  H10K85/50

Abstract: A perovskite photoactive composite layer, a preparation method thereof, and a perovskite solar cell comprising the same are provided. The perovskite photoactive composite layer can induce smooth charge transport when applied to a solar cell by means of a two-dimensional perovskite photoactive layer grown by recrystallization in the vertical direction, and can passivate surface defects in two-dimensional perovskites and prevent the release of halogen atoms through effective passivation by means of a passivation layer consisting of an organic monomolecular compound containing zwitterions, thereby improving the photostability of the solar cell.

公开(公告)号：US20240378701A1

公开(公告)日：2024-11-14

申请号：US18652278

申请日：2024-05-01

Applicant: SAMSUNG ELECTRONICS CO., LTD. ,  Gwangju Institute of Science and Technology

Inventor： Seokpyo HONG ,  Bochang MOON ,  Nahyup KANG ,  Hajin CHOI ,  Inwoo HA

IPC: G06T5/60 ,  G06T5/70 ,  G06T15/00 ,  G06T15/20

Abstract: An electronic device and operating method for removing a noise of an image are disclosed. A method of operating an electronic device includes generating a common input including an initial image and geometry buffer (G-buffer) images rendered according to a view point of a current frame, generating a third input and a fourth input based on a first input and a second input, determining a bandwidth for filtering noise of the initial image based on the common input and one of the third input and the fourth input, and outputting a target image obtained by removing noise from the initial image of the current frame based on the common input, the third input, the fourth input, and the bandwidth.

公开(公告)号：US12140562B2

公开(公告)日：2024-11-12

申请号：US17380703

申请日：2021-07-20

Applicant: GIST (Gwangju Institute of Science and Technology)

Inventor： Sung Yang ,  Taek Eon Jeong ,  Jae Hyeon Ahn ,  Yu Gyung Jung

IPC: G01N27/447

Abstract: The present disclosure relates to a microorganism detection apparatus using a dielectrophoresis (DEP) force. A microorganism detection apparatus according to one embodiment of the present disclosure may include a detection unit that detects microbial particles using a DEP force corresponding to latex particles combined with the microbial particles.

公开(公告)号：US20240354565A1

公开(公告)日：2024-10-24

申请号：US18434417

申请日：2024-02-06

Applicant: GIST(Gwangju Institute of Science and Technology)

Inventor： Hong Kook KIM ,  Geon Woo LEE

IPC: G06N3/08

CPC classification number: G06N3/08

Abstract: The present disclosure relates to a method for generating a personalized HRTF using a neural network model having a one-to-many structure. A method for generating a personalized HRTF according to an embodiment of the present disclosure includes: training a neural network model using multi-angle Head-Related Transfer Functions (HRTF) labeled to body information of a learning object; and obtaining multi-angle HRTFs at a time by inputting body information of a target user into the trained neural network model.

公开(公告)号：US20240330648A1

公开(公告)日：2024-10-03

申请号：US18596994

申请日：2024-03-06

Applicant: Gwangju Institute of Science and Technology

Inventor： Hee Jun JUNG ,  Kang Il KIM ,  Do Yeon KIM

IPC: G06N3/042 ,  G06N3/082 ,  G06N3/096

CPC classification number: G06N3/042 ,  G06N3/082 ,  G06N3/096

Abstract: A method for training a student network including at least one or more of a transformer neural network by using knowledge distillation in a teacher network including at least one or more of the transformer neural network is disclosed. The method includes: pre-training the teacher network using a training data and fine tuning the trained teacher network; copying a weight parameter of a bottom layer of the teacher network to the student network; and performing the knowledge distillation to the student network through the fine-tuned teacher network. The performing the knowledge distillation includes: extracting a feature structure from the result value of a layer of the fine-tuned teacher network; extracting a feature structure from the result value of a layer of the student network; and adjusting the feature structure of the extracted student network based on the feature structure of the extracted teacher network.