Jove
Visualize
Contact Us

Related Experiment Videos

A graph-spectral approach to shape-from-shading.

Antonio Robles-Kelly1, Edwin R Hancock

  • 1University of York, York YO 105DD, UK. arobkell@cs.york.ac.uk

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|January 15, 2005
PubMed
Summary

This study introduces a novel shape-from-shading algorithm using graph-spectral methods for accurate surface reconstruction. The approach integrates surface normals and height data iteratively for improved 3D shape recovery from images.

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Learning From Human Attention for Attribute-Assisted Visual Recognition.

IEEE transactions on pattern analysis and machine intelligence·2024
Same author

The Ihara zeta function as a partition function for network structure characterisation.

Scientific reports·2024
Same author

Retinal vessel segmentation via a Multi-resolution Contextual Network and adversarial learning.

Neural networks : the official journal of the International Neural Network Society·2023
Same author

Contrastive Learning for Joint Normal Estimation and Point Cloud Filtering.

IEEE transactions on visualization and computer graphics·2023
Same author

Regionwise Generative Adversarial Image Inpainting for Large Missing Areas.

IEEE transactions on cybernetics·2022
Same author

Two-Level Graph Neural Network.

IEEE transactions on neural networks and learning systems·2022
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

Area of Science:

  • Computer Vision
  • Computer Graphics
  • Computational Geometry

Background:

  • Shape-from-shading is a challenging problem in computer vision.
  • Existing methods often struggle with noise and complex surface variations.

Purpose of the Study:

  • To develop a novel shape-from-shading algorithm using graph-spectral methods.
  • To improve the accuracy and robustness of 3D surface reconstruction from images.

Main Methods:

  • Characterizing surface normals using a weight matrix based on sectional curvature.
  • Employing graph seriation for surface integration path optimization.
  • Iteratively refining height recovery and surface normal adjustment, ensuring compliance with Lambert's law.

Main Results:

Related Experiment Videos

  • Demonstrated effective 3D surface reconstruction from both synthetic and real-world imagery.
  • The proposed method shows robustness in handling variations in surface normals.
  • Achieved a stable and smoothed reconstructed surface through iterative refinement.

Conclusions:

  • Graph-spectral methods offer a powerful framework for developing advanced shape-from-shading algorithms.
  • The proposed iterative approach enhances the reliability of 3D surface reconstruction.
  • This method advances the field of passive 3D vision and surface analysis.