S-Space College of Engineering/Engineering Practice School (공과대학/대학원) Dept. of Industrial Engineering (산업공학과) Theses (Master's Degree_산업공학과)
Style Transfer Learning using Image Segmentation Awareness
이미지 분할 인식을 이용한 스타일 전이 학습
- 공과대학 산업공학과
- Issue Date
- 서울대학교 대학원
- 학위논문 (석사)-- 서울대학교 대학원 : 공과대학 산업공학과, 2018. 2. 이재욱.
- Recently, there are many studies in image style transfer. Image style transfer method transfers the style to the content image. However, it has a problem that it transfers the same style in different semantics. Thus, we proposed a new method named 'Segment style transfer'. Our method is composed of 4 phases: 'Image Segmentation', 'Segment Matching', 'Segment Style Transfer', and 'Segment Merging'. Our Segment style transfer method improves the existing image style transfer in that it does not make the transferred image heterogeneous and improve the image quality.