ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술대회 Generative Adversarial Network for Identifying Authors of Traced Images
Cited 0 time in scopus Download 2 time Share share facebook twitter linkedin kakaostory
저자
정순철, 김재우, 최윤석, 전형주, 김진서
발행일
202010
출처
International Conference on Information and Communication Technology Convergence (ICTC) 2020, pp.1299-1302
DOI
https://dx.doi.org/10.1109/ICTC49870.2020.9289592
협약과제
20IH3600, 근현대 미술품들의 디지털 데이터 확보 및 과학 기반 미술품 신뢰도 분석 지원 시스템 개발, 김진서
초록
Art forgery is one of the important issues in art, but the general approaches using well-known image datasets are likely to fail to identify real forgeries. This is because the distributions of the well-known image datasets are very different from that of a dataset that includes the pictures created by professional counterfeiters. However, there are few datasets specialized for art forgery detection. In this research, we constructed a dataset for author identification of drawings. To create the drawings, we traced a few original sketches on top of semi-transparent papers. The traced images look more similar to the originals than typical forgery artwork images do, and it implies that it is more difficult to identify the authors of the traced images. We also suggest a new generative adversarial network model for author identification. We approach the problem by maximizing mutual information between an author and his traced artwork image. We compared our results to those obtained by the baseline convolutional neural networks. The experimental results showed that the suggested approach produced more convincing results than other approaches.
키워드
Author Identification, Forgery Detection, Generative Adversarial Network, Mutual Information, Tracing Artworks
KSP 제안 키워드
Author Identification, Convolution neural network(CNN), Image datasets, Network model, Semi-transparent, forgery detection, generative adversarial network, mutual information