EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Frequency–Geometry-Guided Network for Depth Map Super-Resolution

Zhiqiang Feng, Chong Zhang

Sensors · 05.07.2026 · 26 (13) · 4282

DOI: 10.3390/s26134282 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 05.07.2026. Verfasst von Zhiqiang Feng und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Automation, School of Computer Science and Artificial Intelligence, Beijing Technology and Business University, Beijing 100048, China. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Beijing Technology and Business University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Depth super-resolution reconstructs high-resolution (HR) depth maps from low-resolution (LR) inputs with the aid of HR RGB guidance, but RGB edges often do not coincide with true depth discontinuities, causing texture copying and degraded geometric consistency. To address this problem, we propose Frequency–Geometry-Guided Network (FGGNet), a spatial–frequency fusion framework for RGB-guided depth map super-resolution. FGGNet introduces Multi-branch RGB-guided Convolution (MRGConv) to enhance RGB structural representations, a Geometry Prior-guided Fusion Module (GPFM) to filter geometrically inconsistent RGB responses using depth-derived priors, and radial complex spectral loss (RCSL) to emphasize boundary-related high-frequency components in the complex spectral domain. Experiments on NYU v2, Middlebury, Lu, and RGB-D-D show that FGGNet achieves competitive or superior reconstruction accuracy under synthetic and real-world degradation settings. Under the ×16 setting, FGGNet reduces RMSE by 13.7%, 22.8%, 18.5%, and 11.4% on the four datasets, respectively, compared with the average RMSE of five representative state-of-the-art methods. These results validate the effectiveness of combining geometric prior filtering with frequency-domain supervision for reliable depth reconstruction.

Schlagwörter

Stichwörter der Autor:innen

RGB-guided depth restorationdepth map super-resolutionfrequency-domain supervisiongeometric priorspatial frequency domain

Autor:innen und Institutionen

  • Department of Automation, School of Computer Science and Artificial Intelligence, Beijing Technology and Business University, Beijing 100048, China.

Förderung

  • National Natural Science Foundation of China
  • Beijing Technology and Business University

Zitieren

APA

Feng, Z., & Zhang, C. (2026). Frequency–Geometry-Guided Network for Depth Map Super-Resolution. Sensors, 26(13), 4282. https://doi.org/10.3390/s26134282

Vancouver

Feng Z, Zhang C. Frequency–Geometry-Guided Network for Depth Map Super-Resolution. Sensors. 2026;26(13):4282. doi:10.3390/s26134282

BibTeX
@article{Feng2026,
  author = {Feng, Zhiqiang and Zhang, Chong},
  title = {Frequency–Geometry-Guided Network for Depth Map Super-Resolution},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {13},
  pages = {4282},
  doi = {10.3390/s26134282}
}
Kostenlos starten →