인문학
사회과학
자연과학
공학
의약학
농수해양학
예술체육학
복합학
DBpia AI
AI 에이전트
AI 아이디어
AI 리더
AI 뷰어
Citeasy
크롬 서지관리 다운로드
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.
이용수
초록· 키워드
Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have achieved notable success in photorealistic 3D reconstruction and novel view synthesis but rely heavily on high-quality multi-view images, which limits their robustness under real-world degradations such as noise, blur, low-resolution (LR), and weather artifacts. To address this, 3D Low-Level Vision (3D LLV) extends classical 2D restoration tasks such as deblurring and weather degradation removal into the 3D domain. This survey formalizes the problem of degradation-aware rendering and outlines key challenges related to spatio-temporal consistency and ill-posed optimization. It categorizes recent approaches that integrate LLV into neural rendering frameworks and examines their applicability to domains including autonomous driving, AR/VR, and robotics. By reviewing representative methods, datasets, and evaluation protocols, this work identifies 3D LLV as a fundamental direction for robust 3D scene reconstruction under real-world conditions.
상세정보 수정요청해당 페이지 내 제목·저자·목차·페이지정보가 잘못된 경우 알려주세요!
목차
- Abstract
- I. Introduction
- II. Problem Definition and Challenges
- III. Low-Level Vision for Robust Rendering
- IV. Dataset and Metrics
- V. Conclusion
- 참고문헌
참고문헌
참고문헌 신청최근 본 자료
UCI(KEPA) : I410-151-25-02-093839670
