公开(公告)号：US20240211510A1

公开(公告)日：2024-06-27

申请号：US18396400

申请日：2023-12-26

Applicant: POSTECH RESEARCH AND BUSINESS DEVELOPMENT FOUNDATION

Inventor： Min Su CHO ,  Da Hyun KANG

IPC: G06F16/583 ,  G06F16/532 ,  G06F16/55 ,  G06T7/10

CPC classification number: G06F16/583 ,  G06F16/532 ,  G06F16/55 ,  G06T7/10

Abstract: A few-shot learning method according to the present disclosure includes obtaining an image and a segmentation prediction for the image and learning, when a query image is provided with a pre-learned model, the model based on the image and a segmentation index to simultaneously perform classification and segmentation of a specific region from the query image, in which the model includes an attentive squeeze network (ASNet), does not perform the classification when an object with low relevance exists in the query image, and performs the classification and segmentation when an object with high relevance exists in the query image.