ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Journal Article AI 기반 동적 공간영상 생성 기술 동향
Cited - time in scopus Download 6 time Share share facebook twitter linkedin kakaostory
Authors
임한신, 추현곤
Issue Date
2025-04
Citation
전자통신동향분석, v.40, no.2, pp.1-10
ISSN
1225-6455
Publisher
한국전자통신연구원
Language
Korean
Type
Journal Article
DOI
https://dx.doi.org/10.22648/ETRI.2025.J.400201
Abstract
Estimating the 3D geometric structure of a scene from images and generating accurate 3D models has been a long-standing research topic and remains an active area of study. However, the process of creating accurate 3D models requires relatively complex steps and achieving the desired quality of a 3D model is often challenging. Considering the complexity and difficulty of generating 3D models, technologies that create novel view images from scene images without precise 3D model information have seen consistent development. Recently, advancements in novel view synthesis technologies that generate images from given scene data have accelerated significantly, particularly after demonstrating the potential of combining these methods with AI technologies to produce higher-quality results. Among AI-based novel view synthesis approaches, AI-based dynamic scene reconstruction and rendering have received significant attention. This study explored recent trends in AI-based novel view synthesis for dynamic scenes.
Keyword
AI, NeRF, Novel View Synthesis, 3DGS, 4DGS, 동적공간영상
KSP Keywords
3D Model, Active area, Geometric structure, Novel view synthesis, Recent trends, Scene Reconstruction, Scene images, dynamic scenes, model information
This work is distributed under the term of Korea Open Government License (KOGL)
(Type 4: : Type 1 + Commercial Use Prohibition+Change Prohibition)
Type 4: