EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Confidence-Guided Fusion for Self-Supervised Monocular Depth Estimation in Endoscopy

Shuang Li, Hongbo Wang, Zhaoxu Hu, Tian Chu, Yingping Li iD, Liang Zhao

Sensors · 25.06.2026 · 26 (13) · 4033

DOI: 10.3390/s26134033 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 25.06.2026. Verfasst von Shuang Li und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Information and Artificial Intelligence, Xi'an University of Architecture and Technology Huaqing College, Xi'an 710043, China. Hauptthemen laut MEDLINE: Endoscopy, Depth Perception. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Youth Innovation Team of Shaanxi Universities of Xi’an University of Architecture and Technology Huaqing College, School-level Special Project on Artificial Intelligence Empowering Education and Teaching Reform of Xi’an University of Architecture and Technology Huaqing Col….

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate monocular depth estimation (MDE) is a foundational task in endoscopic surgery, critical for augmenting depth perception and aiding surgical navigation. While diffusion-based and discriminative depth estimators demonstrate complementary strengths, they also exhibit asymmetric errors: discriminative models yield precise geometric boundaries but struggle in homogeneous or saturated areas, whereas diffusion models recover fine textures at the cost of occasional structural incoherence. To systematically exploit this complementarity, we present CoDepth, a novel framework that leverages confidence-guided fusion to harmonize the outputs of these heterogeneous estimators. Its core components include a complementary map extractor that identifies structured disparity disagreements, a cross-attention module for context-aware feature integration, and a probabilistic confidence network that generates spatially adaptive fusion weights. Extensive evaluations on the SCARED dataset show that CoDepth achieves improved overall performance relative to strong single-model baselines, with the most consistent gains observed in Abs Rel and δ-based accuracy, while changes in some other error metrics are more modest. Furthermore, CoDepth exhibits encouraging cross-domain generalization. When a model trained on SCARED is directly evaluated on SERV-CT, Hamlyn, and C3VD without fine-tuning, it achieves competitive performance and improves several key metrics across datasets. The framework also demonstrates enhanced robustness against common synthetic corruptions like low-light conditions, Gaussian noise, and impulse noise, underscoring its practical utility in complex clinical settings. These results suggest that confidence-guided complementary fusion provides a practical integration-level paradigm for combining heterogeneous endoscopic depth estimators.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

EndoscopyDepth PerceptionHumansAlgorithms

Stichwörter der Autor:innen

confidence-guided fusionendoscopymonocular depth estimationself-supervised learning

Autor:innen und Institutionen

  • School of Information and Artificial Intelligence, Xi'an University of Architecture and Technology Huaqing College, Xi'an 710043, China.
  • School of Mechanical and Electrical Engineering, Xi'an University of Architecture and Technology Huaqing College, Xi'an 710043, China.
  • School of Artificial Intelligence, Xidian University, Xi'an 710071, China.
  • College of Information and Control Engineering, Xi'an University of Architecture and Technology, Xi'an 710055, China.

ORCID-Profile: Yingping Li

Förderung

  • Youth Innovation Team of Shaanxi Universities of Xi’an University of Architecture and Technology Huaqing College
  • School-level Special Project on Artificial Intelligence Empowering Education and Teaching Reform of Xi’an University of Architecture and Technology Huaqing Col…

Zitieren

APA

Li, S., Wang, H., Hu, Z., Chu, T., Li, Y., & Zhao, L. (2026). Confidence-Guided Fusion for Self-Supervised Monocular Depth Estimation in Endoscopy. Sensors, 26(13), 4033. https://doi.org/10.3390/s26134033

Vancouver

Li S, Wang H, Hu Z, Chu T, Li Y, Zhao L. Confidence-Guided Fusion for Self-Supervised Monocular Depth Estimation in Endoscopy. Sensors. 2026;26(13):4033. doi:10.3390/s26134033

BibTeX
@article{Li2026,
  author = {Li, Shuang and Wang, Hongbo and Hu, Zhaoxu and Chu, Tian and Li, Yingping and Zhao, Liang},
  title = {Confidence-Guided Fusion for Self-Supervised Monocular Depth Estimation in Endoscopy},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {13},
  pages = {4033},
  doi = {10.3390/s26134033}
}
Kostenlos starten →