no code implementations • 21 Dec 2023 • Soopil Kim, Sion An, Philip Chikontwe, Myeongkyun Kang, Ehsan Adeli, Kilian M. Pohl, Sang Hyun Park
In this study, we introduce a novel component segmentation model for LA detection that leverages a few labeled samples and unlabeled images sharing logical constraints.
no code implementations • CVPR 2022 • Philip Chikontwe, Soopil Kim, Sang Hyun Park
Few-shot classification is a challenging problem that aims to learn a model that can adapt to unseen classes given a few labeled samples.
no code implementations • 18 Oct 2021 • Soopil Kim, Philip Chikontwe, Sang Hyun Park
During inference, query segmentation is predicted using prototypes from both support and unlabeled images including low-level features of the query images.
no code implementations • 21 Apr 2021 • Heejung Park, Gyeong Min Lee, Soopil Kim, Ga Hyung Ryu, Areum Jeong, Sang Hyun Park, Min Sagong
To quickly adapt to various tasks, the meta learner was updated to get close to the center of parameters which are fine-tuned for each registration task.
no code implementations • 19 Nov 2020 • Soopil Kim, Sion An, Philip Chikontwe, Sang Hyun Park
In this paper, we propose a 3D few shot segmentation framework for accurate organ segmentation using limited training samples of the target organ annotation.
no code implementations • 3 Mar 2020 • Sion An, Soopil Kim, Philip Chikontwe, Sang Hyun Park
In addition to the unified learning of feature similarity and a few shot classifier, our method leads to emphasize informative features in support data relevant to the query data, which generalizes better on unseen subjects.