Depth estimation of a single image presents a classic problem for computer vision, and is important for the 3D reconstruction of scenes, augmented reality, and object detection. At present, most researchers are beginning to focus on unsupervised monocular depth estimation. This paper proposes solutions to the current depth estimation problem. These solutions include a monocular depth estimation method based on uncertainty analysis, which solves the problem in which a neural network has strong expressive ability but cannot evaluate the reliability of an output result. In addition, this paper proposes a photometric loss function based on the Retinex algorithm, which solves the problem of pulling around pixels due to the presence of moving objects. We objectively compare our method to current mainstream monocular depth estimation methods and obtain satisfactory results.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7570747 | PMC |
| http://dx.doi.org/10.3390/s20185389 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Association Between Patient Characteristics and the Depth of Microvascular Penetration Into the Adult Human Meniscus.
Am J Sports Med
January 2025
Department of Orthopaedics, Isala Hospital, Zwolle, The Netherlands.
Background: Current knowledge on the microvascular anatomy of adult human menisci is based on cadaveric studies. However, considerable interindividual variation in meniscal microvascularization has been reported in recent studies with small sample sizes.
Purpose: To assess the association between patient characteristics and the depth of microvascularization of the meniscus.
Speckle-illumination spatial frequency domain imaging with a stereo laparoscope for profile-corrected optical property mapping.
J Biomed Opt
January 2025
The Johns Hopkins University, Department of Biomedical Engineering, Baltimore, Maryland, United States.
Significance: Laparoscopic surgery presents challenges in localizing oncological margins due to poor contrast between healthy and malignant tissues. Optical properties can uniquely identify various tissue types and disease states with high sensitivity and specificity, making it a promising tool for surgical guidance. Although spatial frequency domain imaging (SFDI) effectively measures quantitative optical properties, its deployment in laparoscopy is challenging due to the constrained imaging environment.
View Article and Find Full Text PDFLithospheric strike-slip faulting in central Tibet since 35-32 Ma and implications for the incipient Asian extrusional tectonics.
Natl Sci Rev
February 2025
SinoProbe Laboratory, Key Laboratory of Continental Dynamics of Ministry of Natural Resources, Institute of Geology, Chinese Academy of Geological Sciences, Beijing 100037, China.
The onset age and depth of the central Tibet strike-slip faults are two still unresolved fundamental issues with regard to the Cenozoic tectonic evolution of central Tibet. Here we present a comprehensive dataset of geochronological, geochemical and structural data on recently discovered en-echelon dykes representing the incipient development of strike-slip faulting from the Lunpola basin in central Tibet. Our results provide evidence for mantle-derived, bimodal magmatism linked to lithospheric-scale strike-slip faulting at 35-32 Ma, and demonstrate that the central Tibet strike-slip faults are at least 20 Ma older than previously estimated (15-8 Ma).
View Article and Find Full Text PDFFluorescence molecular imaging-guided photodynamic therapy for early breast cancer in the prone position: feasibility evaluation with Monte Carlo simulations.
Photodiagnosis Photodyn Ther
January 2025
Maebashi-Institute of Technology, Systems Life Engineering, Gunma, 371-0816 Japan. Electronic address:
Introduction: The successful diagnosis and treatment of early-stage breast cancer enhances the quality of life of patients. As a promising alternative to recently developed magnetic resonance imaging-guided radiotherapy, we proposed fluorescence molecular imaging-guided photodynamic therapy (FMI-guided PDT), which requires no expensive equipment. In the FMI simulations, ICG-C11 which has emission peaks at near-infrared wavelengths was used as the FMI agent.
View Article and Find Full Text PDFXIS-PM: A daily spatiotemporal machine-learning model for PM in the contiguous United States.
Environ Res
January 2025
Department of Environmental Medicine and Public Health, Icahn School of Medicine at Mount Sinai, New York, NY, USA; The Department of Geography and Environmental Development, Ben-Gurion University of the Negev, Beer Sheva, Israel.
Air-pollution monitoring is sparse across most of the United States, so geostatistical models are important for reconstructing concentrations of fine particulate air pollution (PM) for use in health studies. We present XGBoost-IDW Synthesis (XIS), a daily high-resolution PM machine-learning model covering the contiguous US from 2003 through 2023. XIS uses aerosol optical depth from satellites and a parsimonious set of additional predictors to make predictions at arbitrary points, capturing near-roadway gradients and allowing the estimation of address-level exposures.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!