60‐3: Distinguished Paper: A Method for Generating New Viewpoints in Monocular Images Based on Diffusion Model
SID Symposium Digest of Technical Papers · 2026-06 · 57 (1) · 843-846
DOI: 10.1002/sdtp.70001_230 ↗Kurz gefasst
Fachartikel in der Zeitschrift SID Symposium Digest of Technical Papers, online erschienen am 11.09.2026. Verfasst von Yingdong Gu und 3 weiteren Autor:innen. Die Arbeit nennt 21 Literaturangaben.
Abstract (Originalsprache)
Deep learning shows significant promise in 3D content generation; however, current methods often lack robustness in complex scenes, produce low‐resolution outputs, and yield low Mean Opinion Scores (MOS). This paper presents a diffusion‐based approach that frames novel view synthesis as an image restoration problem. Our method enables robust 3D content generation at 2K resolution, maintains fine texture details up to 16K, and attains an MOS of 3.83 on 4K two‐view 3D displays.
Zitieren
APA
Gu, Y., Li, Y., An, Z., & Jiang, X. (2026). 60‐3: Distinguished Paper: A Method for Generating New Viewpoints in Monocular Images Based on Diffusion Model. SID Symposium Digest of Technical Papers, 57(1), 843-846. https://doi.org/10.1002/sdtp.70001_230
Vancouver
Gu Y, Li Y, An Z, Jiang X. 60‐3: Distinguished Paper: A Method for Generating New Viewpoints in Monocular Images Based on Diffusion Model. SID Symposium Digest of Technical Papers. 2026;57(1):843-846. doi:10.1002/sdtp.70001_230
BibTeX
@article{Gu2026,
author = {Gu, Yingdong and Li, Yingjie and An, Zhanfu and Jiang, Xingqun},
title = {60‐3: Distinguished Paper: A Method for Generating New Viewpoints in Monocular Images Based on Diffusion Model},
journal = {SID Symposium Digest of Technical Papers},
year = {2026},
volume = {57},
number = {1},
pages = {843--846},
doi = {10.1002/sdtp.70001_230}
}
EUVIMED