KIT | KIT-Bibliothek | Impressum | Datenschutz

A cGAN-based network for depth estimation from bronchoscopic images

Guo, Lu ORCID iD icon 1; Nahm, Werner 1
1 Institut für Biomedizinische Technik (IBT), Karlsruher Institut für Technologie (KIT)

Abstract:

Purpose: Depth estimation is the basis of 3D reconstruction of airway structure from 2D bronchoscopic scenes, which can be further used to develop a vision-based bronchoscopic navigation system. This work aims to improve the performance of depth estimation directly from bronchoscopic images by training a depth estimation network on both synthetic and real datasets.
Methods: We propose a cGAN-based network Bronchoscopic-Depth-GAN (BronchoDep-GAN) to estimate depth from bronchoscopic images by translating bronchoscopic images into depth maps. The network is trained in a supervised way learning from synthetic textured bronchoscopic image-depth pairs and virtual bronchoscopic image-depth pairs, and simultaneously, also in an unsupervised way learning from unpaired real bronchoscopic images and depth maps to adapt the model to real bronchoscopic scenes.
Results: Our method is tested on both synthetic data and real data. However, the tests on real data are only qualitative, as no ground truth is available. The results show that our network obtains better accuracy in all cases in estimating depth from bronchoscopic images compared to the well-known cGANs pix2pix.
... mehr


Verlagsausgabe §
DOI: 10.5445/IR/1000162221
Veröffentlicht am 14.09.2023
Originalveröffentlichung
DOI: 10.1007/s11548-023-02978-z
Scopus
Zitationen: 1
Web of Science
Zitationen: 2
Dimensions
Zitationen: 1
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Biomedizinische Technik (IBT)
Publikationstyp Zeitschriftenaufsatz
Publikationsjahr 2023
Sprache Englisch
Identifikator ISSN: 1861-6429
KITopen-ID: 1000162221
Erschienen in International Journal of Computer Assisted Radiology and Surgery
Verlag Springer-Verlag
Band 19
Heft 1
Seiten 33–36
Vorab online veröffentlicht am 10.08.2023
Schlagwörter Depth estimation, Conditional GANs, Bronchoscopy, Image-guided surgery
Nachgewiesen in Dimensions
Scopus
Web of Science
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page