지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학술저널
- 저자정보
- 발행연도
- 2023.8
- 수록면
- 535 - 545 (11page)
- DOI
- 10.9728/jcc.2023.06.5.1.535
이용수
초록· 키워드
오류제보하기
People can completely absorb a scene through the emerging technique of 3D imaging. In this study, we introduce a method to generate a 3D view from a single input image, enabling captivating 3D scenes using Korean historic images. We present a layered illustration approach for viewpoint generation that incorporates color and depth features in areas where the original image is obscured. Our technique involves a multilayered Depth Illustration that achieves precise spatial interaction for a 3D perspective. Additionally, we utilize a deep learning-based coloring approach that enhances regional shades and depth information in obscured areas while considering spatial information. By employing this method, we create 3D images from a single image input with fewer parameters compared to other methods. The generated 3D images can be effectively displayed with motion using conventional display engines. We evaluate the efficiency of our approach on a range of challenging images, demonstrating its effectiveness in conserving RAM and minimizing delays at high rates. This approach simplifies the process of 3D imaging by enabling the generation of 3D views from a single image, which can be used to create virtual reality (VR) and augmented reality (AR) experiences as well as can be used to reconstruct ancient artifacts and structures by estimating the depth of features in an image.
목차
Abstract
1 Introduction
2 Related Work
3 Method
4 Results
5 Conclusion
References
참고문헌 (0)
참고문헌 신청이 논문의 저자 정보
이 논문과 함께 이용한 논문
최근 본 자료
댓글(0)
0