Meta Pseudo Labels 기반 딥페이크 영상 검출
Meta Pseudo Labels Based Deepfake Video Detection
- 발행
- 2024 등록KCI API에는 발행일 항목이 없어 원문 등록일을 쓰고 있습니다. 실측으로 발행보다 최대 6년 늦습니다 — 재수집하면 발행연월로 바뀝니다.
- 소속·발행
- 동아대학교
- 출처
- 국내 KCI
- DOI
- 10.9717/kmms.2024.27.1.009
- 원문
- 원문 보기 ↗
개념
키워드
Deepfake Detection, Meta Learning, Meta Pseudo Labels, Deepfake Unknown Domain, Generative Misuse, Deepfake Detection, Meta Learning, Meta Pseudo Labels, Deepfake Unknown Domain, Generative Misuse
초록
Recently, there has been considerable research on deepfake detection. However, most existing meth ods face challenges in adapting to the advancements in new generative models within unknown domains. In this paper, our objective is to detect deepfake videos in unknown domains using unlabeled data. Specifically, our proposed approach employs Meta Pseudo Labels (MPL), allowing the model to be trained on unlabeled images. MPL involves the simultaneous training of both a Teacher model and a Student model, where the Teacher model generates Pseudo Labels utilized to train the Student model.
 This method aims to enhance the adaptability and robustness of deepfake detection systems against emerging unknown domains. The experimental results demonstrate an improvement of 1.91% and 1.87% in ACC and AUROC, respectively, for the known domain. Similarly, in the unknown domain, there is an enhancement of 1.59% in ACC and 1.29% in AUROC.