학술행사
- 발표자 현윤석(인하대학교)
- 조직위원 산업수학혁신센터
- 기간 2023-12-14 ~ 2023-12-14
- 장소 판교 테크노밸리 산업수학혁신센터 세미나실
- 주최 산업수학혁신센터
-
일시: 2023년 12월 14일(목), 14:00-16:00
-
장소: 판교 테크노밸리 산업수학혁신센터 세미나실
- 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
- 무료주차는 2시간 지원됩니다.
-
발표자: 현윤석(인하대학교)
-
주요내용: Siamese Networks in Computer Vision
In the realm of computer vision, the Siamese network has emerged as a powerful paradigm, revolutionizing image processing and analysis. The primary objective of a Siamese network is to learn a similarity metric between pairs of input samples. The Siamese architecture, characterized by its unique structure of shared weights between twin networks, has proven to be highly effective in tasks such as image similarity, object tracking, depth estimation, and facial recognition. Recently, in the context of self-supervised learning in computer vision, Siamese networks are often employed to train models without relying on traditional labeled datasets. Instead, they exploit the inherent structure or relationships within the data to learn meaningful representations. This seminar delves into the intricacies of Siamese networks and explores their application in various computer vision tasks, including recent works on self-supervised representation learning and other interesting topics.
*유튜브 스트리밍 예정입니다.
-
일시: 2023년 12월 14일(목), 14:00-16:00
-
장소: 판교 테크노밸리 산업수학혁신센터 세미나실
- 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
- 무료주차는 2시간 지원됩니다.
-
발표자: 현윤석(인하대학교)
-
주요내용: Siamese Networks in Computer Vision
In the realm of computer vision, the Siamese network has emerged as a powerful paradigm, revolutionizing image processing and analysis. The primary objective of a Siamese network is to learn a similarity metric between pairs of input samples. The Siamese architecture, characterized by its unique structure of shared weights between twin networks, has proven to be highly effective in tasks such as image similarity, object tracking, depth estimation, and facial recognition. Recently, in the context of self-supervised learning in computer vision, Siamese networks are often employed to train models without relying on traditional labeled datasets. Instead, they exploit the inherent structure or relationships within the data to learn meaningful representations. This seminar delves into the intricacies of Siamese networks and explores their application in various computer vision tasks, including recent works on self-supervised representation learning and other interesting topics.
