이동우연구실의 신OO 학생이 작년 말 세계 최고 권위의 국제학술지인 IEEE(Institute of Electrical and Electronics Engineers)의
INTERNATIONAL CONFERENCE OF IMAGE PROCESSING(ICIP)에 단독 연구자로 제출한 영어 논문이 accept되어서 2024년 9월에 실리게 되었습니다.
이런 결과는 명문 자사고나, 영재학교에서도 굉장히 드문 성과이며, 학년이 없이 연구 중심의 개별교육과정을 운영하는 마이폴학교에서나 가능한 일입니다.
많이 축하해 주세요.
논문 제목 :
Mask-based Invisible Backdoor Attacks on Object Detection
OO Shin
Deep learning models have achieved unprecedented performance in the domain of object detection, resulting in breakthroughs in areas such as autonomous driving and security. However, deep learning models are vulnerable to backdoor attacks. These attacks prompt models to behave similarly to standard models without a trigger; however, they act maliciously upon detecting a predefined trigger. Despite extensive research on backdoor attacks in image classification, their application to object detection remains relatively underexplored. Given the widespread application of object detection in critical real-world scenarios, the sensitivity and potential impact of these vulnerabilities cannot be overstated. In this study, we propose an effective invisible backdoor attack on object detection utilizing a mask-based approach. Three distinct attack scenarios were explored for object detection: object disappearance, object misclassification, and object generation attack. Through extensive experiments, we comprehensively examined the effectiveness of these attacks and tested certain defense methods to determine effective countermeasures. Code will be available at this https URL
IEEE : (Institute of Electrical and Electronics Engineers)
INTERNATIONAL CONFERENCE OF IMAGE PROCESSING (ICIP) 2024
The 7-PAGE ARTICLE is expected to be published in September, 2024
