公开(公告)号：US20250124598A1

公开(公告)日：2025-04-17

申请号：US18795557

申请日：2024-08-06

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Moon-Ki BACK ,  Jang-Hee YOO ,  Cheol-Hwan YOO ,  Byung-Ok HAN

IPC: G06T7/73 ,  G06V10/776 ,  G06V10/82

Abstract: Disclosed herein are a deep neural network (DNN) learning method for generalizing appearance-based gaze estimation and an apparatus for the same. The deep neural network (DNN) learning method includes creating multiple augmented images based on an original image, inputting the multiple augmented images to a DNN to output a gaze estimation value, calculating a total loss between a gaze ground truth of the original image and the gaze estimation value through gaze consistency regularization (GCR) using a spherical gaze distance (SGD), and updating parameters of the DNN by backpropagation of the total loss.

公开(公告)号：US20250078568A1

公开(公告)日：2025-03-06

申请号：US18502336

申请日：2023-11-06

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Byung-Ok HAN ,  Moon-Ki BACK ,  Jang-Hee YOO ,  Cheol-Hwan YOO

IPC: G06V40/16

Abstract: Disclosed herein are an emotion recognition method and method based on context information. The emotion recognition method includes detecting information corresponding to an emotion recognition subject from an input image, extracting a recognition subject feature based on the information corresponding to the emotion recognition subject, extracting a context feature based on the input image, storing the recognition subject feature and the context feature in a short-term memory, and storing the context feature in a long-term memory.

公开(公告)号：US20240407685A1

公开(公告)日：2024-12-12

申请号：US18623224

申请日：2024-04-01

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Byung-Ok HAN ,  Moon-Ki BACK ,  Jang-Hee YOO ,  Cheol-Hwan YOO

IPC: A61B5/16 ,  A61B5/00 ,  G06V10/70 ,  G06V10/80 ,  G06V20/40 ,  G06V40/16 ,  G06V40/18 ,  G06V40/20 ,  G10L17/02 ,  G10L25/57 ,  G10L25/66 ,  G16H50/20

Abstract: Disclosed herein is a method for supporting Autism Spectrum Disorder (ASD) diagnosis based Artificial Intelligence (AI). The method includes extracting a detection area and voice corresponding to an inspector from an input video, extracting a detection area and voice corresponding to an assessment subject from the input video, extracting a feature of the inspector and a feature of the assessment subject, and extracting an interaction feature using the feature of the inspector and the feature of the assessment subject.

公开(公告)号：US20250125049A1

公开(公告)日：2025-04-17

申请号：US18795543

申请日：2024-08-06

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Cheol-Hwan YOO ,  Moon-Ki BACK ,  Jang-Hee YOO ,  Byung-Ok HAN

IPC: G16H50/20 ,  G16H30/40

Abstract: Disclosed herein is a method for assisting in Autism Spectrum Disorder (ASD) diagnosis. The method includes transmitting social-interaction-inducing content, receiving input images containing a response of an assessment subject to the social-interaction-inducing content, receiving an ASD diagnosis result for a preset number of input images, among the received input images, and outputting a diagnostic assistive result for the received input images using ASD diagnosis input for the preset number of input images and a pretrained global ASD diagnosis model.

公开(公告)号：US20240415413A1

公开(公告)日：2024-12-19

申请号：US18660379

申请日：2024-05-10

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Cheol-Hwan YOO ,  Moon-Ki BACK ,  Jang-Hee YOO ,  Byung-Ok HAN

IPC: A61B5/11

Abstract: Disclosed herein is a method for stereotyped behavior detection for supporting diagnosis of Autism Spectrum Disorder (ASD). The method includes detecting a target object to be assessed in an input video, detecting a section in which a periodic behavior occurs using an image sequence of the target object, and classifying a stereotyped behavior in the section in which the periodic behavior occurs.