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 Experiment Videos

Stereo matching with linear superposition of layers.

Yanghai Tsin1, Sing Bing Kang, Richard Szeliski

  • 1Siemens Corporate Research, 755 College Road East, Princeton, NJ 08540, USA. yanghai.tsin@siemens.com

IEEE Transactions on Pattern Analysis and Machine Intelligence
|February 14, 2006
PubMed
Summary

This study introduces a novel stereo vision algorithm to accurately estimate depth and color in images with non-Lambertian effects. The method overcomes limitations of traditional techniques by analyzing layered image data.

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

Polarimetric Helmholtz Stereopsis.

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

Personalized Exposure Control Using Adaptive Metering and Reinforcement Learning.

IEEE transactions on visualization and computer graphics·2018
Same author

Hyperspectral Light Field Stereo Matching.

IEEE transactions on pattern analysis and machine intelligence·2018
Same author

Content Aware Image Pre-Compensation.

IEEE transactions on pattern analysis and machine intelligence·2018
Same author

Semantic-Driven Generation of Hyperlapse from 360 Degree Video.

IEEE transactions on visualization and computer graphics·2017
Same author

Guest Editorial: Special Section on CVPR 2013.

IEEE transactions on pattern analysis and machine intelligence·2016

Area of Science:

  • Computer Vision
  • Photorealistic Rendering
  • 3D Reconstruction

Background:

  • Traditional stereo vision algorithms struggle with non-Lambertian effects, limiting depth recovery.
  • Non-Lambertian effects, modeled as additive layer superposition, complicate direct color matching.
  • Accurate depth estimation is crucial for 3D scene understanding and augmented reality.

Purpose of the Study:

  • To develop a robust stereo matching algorithm for non-Lambertian image scenarios.
  • To enable accurate estimation of both depth and color for layered image components.
  • To provide a solution for challenging 3D reconstruction tasks where traditional methods fail.

Main Methods:

  • A nested plane sweep approach enumerates depth hypotheses for multiple layers.

Related Experiment Videos

  • Spatial-temporal differencing is employed for matching depth hypotheses.
  • Graph cut optimization and a convergent iterative color update refine estimates.

    • Main Results:

    • The algorithm successfully recovers depth and color for layered image components.
    • Demonstrated effectiveness on both synthetic and real-world image sequences.
    • Outperforms traditional methods in the presence of non-Lambertian effects.

    Conclusions:

    • The proposed method effectively addresses stereo matching challenges posed by non-Lambertian effects.
    • Accurate depth and color recovery is achievable even in complex visual conditions.
    • This work advances the capabilities of 3D reconstruction and computer vision.