상세정보
학술대회
Super Resolution Network for Artistic Paintings
- 발행일
- 202210
- 출처
- International Conference on Information and Communication Technology Convergence (ICTC) 2022, pp.1826-1828
- 협약과제
- 22IH1700, 인공지능 기반의 사진/회화 융합 및 확장생성 콘텐츠 제작기술 개발, 유정재
- 초록
- In this paper, we propose a method of increasing resolution in artistic painting using a deep learning network. Existing SR networks have focused on extracting features that contain sufficient information in a low-resolution state. In this paper, we propose a method that can increase the resolution up to 8 times in the artistic painting domain by improving the upsampling method at the end of the network. The proposed method was confirmed to be usefully applicable to existing artistic painting through experiments.
- KSP 제안 키워드
- Deep learning network, Super resolution, deep learning(DL), low resolution