Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Depth Perception and Spatial Vision01:15

Depth Perception and Spatial Vision

812
Depth perception is the ability to perceive objects three-dimensionally. It relies on two types of cues: binocular and monocular. Binocular cues depend on the combination of images from both eyes and how the eyes work together. Since the eyes are in slightly different positions, each eye captures a slightly different image. This disparity between images, known as binocular disparity, helps the brain interpret depth. When the brain compares these images, it determines the distance to an object.
812
Uniform Depth Channel Flow: Problem Solving01:18

Uniform Depth Channel Flow: Problem Solving

107
To calculate the flow rate for a trapezoidal channel, first, identify the bottom width, side slope, and flow depth of the channel. The cross-sectional area (A) corresponding to the depth of flow (y), channel bottom width (B), and side slope (θ) is determined by:Next, calculate the wetted perimeter, which includes the bottom width and the sloped side lengths in contact with the water. Using the values of the cross-sectional area and the wetted perimeter, determine the hydraulic radius by...
107

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Development and validation of a nomogram to predict bronchopulmonary dysplasia in very preterm, very low birth weight infants.

American journal of translational research·2026
Same author

Similarity measure of time series based on Angle-distance Penalized Metric Dynamic Time Warping.

Scientific reports·2025
Same author

Efficient and robust phase unwrapping via virtual gratings and hybrid coding.

Optics letters·2025
Same author

Preliminary investigation into the mechanical implications of excessive trabecularization in the left ventricle.

Quantitative imaging in medicine and surgery·2025
Same author

Interfacial Cooperative Assembly of Surfactants and Opposite Wettability Nanoparticles Stabilizes Water-in-Oil Emulsions at High Temperature.

Langmuir : the ACS journal of surfaces and colloids·2024
Same author

Towards Stabilized Few-Shot Object Detection with Less Forgetting via Sample Normalization.

Sensors (Basel, Switzerland)·2024

Related Experiment Video

Updated: Aug 16, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K

What makes the unsupervised monocular depth estimation (UMDE) model training better.

Xiangtong Wang1, Binbin Liang1, Menglong Yang1

  • 1School of Aeronautics and Astronautics, Sichuan University, Chengdu, China.

Scientific Reports
|December 20, 2022
PubMed
Summary
This summary is machine-generated.

Generating synthetic data for computer vision tasks like depth estimation is crucial. The MineNavi dataset offers an expandable solution, improving model performance and convergence without manual labeling.

More Related Videos

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders
05:49

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders

Published on: November 1, 2024

882
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

603

Related Experiment Videos

Last Updated: Aug 16, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K
Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders
05:49

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders

Published on: November 1, 2024

882
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

603

Area of Science:

  • Computer Vision
  • Deep Learning
  • Robotics

Background:

  • Deep learning models for computer vision, particularly dense estimation tasks like depth estimation, require extensive annotated data.
  • Manual annotation for dense estimation is labor-intensive and often infeasible, limiting dataset scope and hindering research progress.

Purpose of the Study:

  • To introduce a novel synthetic dataset generation method to overcome limitations of manual data annotation in computer vision.
  • To construct and validate the MineNavi dataset for aircraft navigation applications, focusing on depth estimation.
  • To investigate the impact of synthetic data pre-training on deep learning model performance and convergence.

Main Methods:

  • Developed a synthetic dataset generation pipeline to create expandable datasets without manual workforce.
  • Constructed the MineNavi dataset featuring first-person view aircraft video footage with accurate ground truth for depth estimation.
  • Conducted quantitative experiments to evaluate the effectiveness of MineNavi for pre-training depth estimation models.
  • Performed unsupervised monocular depth estimation (UMDE) experiments on MineNavi to analyze factors influencing model training.

Main Results:

  • Pre-training with the MineNavi dataset significantly improved the performance of depth estimation models.
  • MineNavi pre-training accelerated the convergence of models on real-world scene data.
  • Synthetic data from MineNavi demonstrated comparable effectiveness to real-world datasets in deep model training.
  • Experiments revealed the influence of lighting conditions and motion modes within MineNavi on UMDE model training.

Conclusions:

  • Synthetic dataset generation is a viable and efficient approach to address data scarcity in computer vision.
  • The MineNavi dataset provides a valuable resource for advancing depth estimation and unsupervised monocular depth estimation research.
  • Further exploration of factors within synthetic datasets can optimize deep learning model training for specific applications.