EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LapCR-Net: A Lightweight Monocular Depth Estimation Network via Laplacian Residual Reconstruction

Linghao Li, Yingjun Zhao, Kai Qin, Donghua Lu, Huilin Yang, Ximin Wang

Sensors · 04.08.2026 · 26 (15) · 4912

DOI: 10.3390/s26154912 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 04.08.2026. Verfasst von Linghao Li und 5 weiteren Autor:innen. Beteiligte Einrichtung: Beijing Research Institute of Uranium Geology, Beijing 100029, China. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Monocular depth estimation plays a crucial role in applications such as autonomous driving and mobile 3D reconstruction. However, existing lightweight methods are often constrained by limited computational resources and rely on shallow feature representations for direct depth regression. As a result, cross-scale residual information is insufficiently modeled, which limits their ability to preserve structural consistency and recover fine-grained details in complex scenes. To address these challenges, we propose LapCR-Net, a lightweight monocular depth estimation network based on Laplacian residual reconstruction. Specifically, we formulate a progressive Laplacian residual framework that decomposes depth prediction into a coarse-to-fine multi-scale refinement process. To enhance feature representation in the decoder, we introduce a Structure-aware Feature Recalibration (SFR) module and a Depth-guided Convolution Module (DCM), which strengthen spatial semantic correlations and improve residual prediction across scales. Furthermore, we design an uncertainty-driven collaborative refinement strategy to adaptively adjust residual correction strength. By estimating prediction uncertainty, the proposed strategy sharpens object boundaries while suppressing texture artifacts. Extensive experiments on the NYU-Depth V2 and KITTI benchmarks demonstrate that LapCR-Net achieves competitive performance with only 5.4 M parameters. In particular, it shows clear advantages in structural preservation and detail reconstruction, achieving a favorable trade-off between accuracy and computational efficiency.

Schlagwörter

Stichwörter der Autor:innen

Laplacian residual reconstructionlightweight networkmonocular depth estimationresidual learninguncertainty estimation

Autor:innen und Institutionen

  • Beijing Research Institute of Uranium Geology, Beijing 100029, China.
  • National Key Laboratory of Uranium Resources Exploration-Mining and Nuclear Remote Sensing, Beijing 100029, China.

Zitieren

APA

Li, L., Zhao, Y., Qin, K., Lu, D., Yang, H., & Wang, X. (2026). LapCR-Net: A Lightweight Monocular Depth Estimation Network via Laplacian Residual Reconstruction. Sensors, 26(15), 4912. https://doi.org/10.3390/s26154912

Vancouver

Li L, Zhao Y, Qin K, Lu D, Yang H, Wang X. LapCR-Net: A Lightweight Monocular Depth Estimation Network via Laplacian Residual Reconstruction. Sensors. 2026;26(15):4912. doi:10.3390/s26154912

BibTeX
@article{Li2026,
  author = {Li, Linghao and Zhao, Yingjun and Qin, Kai and Lu, Donghua and Yang, Huilin and Wang, Ximin},
  title = {LapCR-Net: A Lightweight Monocular Depth Estimation Network via Laplacian Residual Reconstruction},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {15},
  pages = {4912},
  doi = {10.3390/s26154912}
}
Kostenlos starten →