EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Comprehensive Survey on Diffusion Model-Driven 3D Reconstruction: Methods, Datasets, and Prospects

Qianwen Yao, Caixia Liu, Jiulin Liang, Haisheng Li, Xiaoqun Wu, Xiaoqiang Teng

ISPRS International Journal of Geo-Information · 01.07.2026 · 15 (7) · 293

DOI: 10.3390/ijgi15070293 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ISPRS International Journal of Geo-Information, online erschienen am 01.07.2026. Verfasst von Qianwen Yao und 5 weiteren Autor:innen. Die Arbeit nennt 146 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Beijing Natural Science Foundation, R&D Program of the Beijing Municipal Education Commission.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Three-dimensional (3D) reconstruction serves as a key technology bridging the real and digital worlds, with broad application in remote sensing, autonomous driving, and robotics. In recent years, its technical paradigm has shifted from geometry-based methods to data-driven approaches, with diffusion models emerging as a major driving force due to their stable training process, strong ability to learn priors from large-scale datasets, and excellent controllability over outputs. Despite the proliferation of diverse architectures, a systematic analysis and comparison of these methods remains absent. In this paper, we present a comprehensive review of diffusion-based 3D reconstruction methods. Based on the space where diffusion operates, we first categorize existing approaches into four types—image diffusion, latent diffusion, 3D diffusion, and hybrid diffusion—and we provide a detailed analysis of their methodologies. We then summarize commonly used 3D datasets and provide a comparative evaluation of these methods across three dimensions: reconstruction accuracy, computational efficiency, and generalization capabilities. Finally, we discuss future developments of diffusion-based 3D reconstruction methods.

Förderung

  • National Natural Science Foundation of China
  • Beijing Natural Science Foundation
  • R&D Program of the Beijing Municipal Education Commission
  • Research Foundation for Youth Scholars of Beijing Technology and Business University

Zitieren

APA

Yao, Q., Liu, C., Liang, J., Li, H., Wu, X., & Teng, X. (2026). A Comprehensive Survey on Diffusion Model-Driven 3D Reconstruction: Methods, Datasets, and Prospects. ISPRS International Journal of Geo-Information, 15(7), 293. https://doi.org/10.3390/ijgi15070293

Vancouver

Yao Q, Liu C, Liang J, Li H, Wu X, Teng X. A Comprehensive Survey on Diffusion Model-Driven 3D Reconstruction: Methods, Datasets, and Prospects. ISPRS International Journal of Geo-Information. 2026;15(7):293. doi:10.3390/ijgi15070293

BibTeX
@article{Yao2026,
  author = {Yao, Qianwen and Liu, Caixia and Liang, Jiulin and Li, Haisheng and Wu, Xiaoqun and Teng, Xiaoqiang},
  title = {A Comprehensive Survey on Diffusion Model-Driven 3D Reconstruction: Methods, Datasets, and Prospects},
  journal = {ISPRS International Journal of Geo-Information},
  year = {2026},
  volume = {15},
  number = {7},
  pages = {293},
  doi = {10.3390/ijgi15070293}
}
Kostenlos starten →