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

Adaptive support-weight approach for correspondence search.

Kuk-Jin Yoon1, In So Kweon

  • 1Department of Electrical Engineering and Computer Science, KAIST, 373-1, Guseong-dong, Yuseong-gu, Daejeon, Korea. kjyoon@rcv.kaist.ac.kr

IEEE Transactions on Pattern Analysis and Machine Intelligence
|March 29, 2006
PubMed
Summary

This study introduces a novel window-based approach for image correspondence search. By adjusting pixel weights based on color and proximity, the method significantly reduces ambiguity and improves accuracy on stereo benchmarks.

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

Toward Enhanced Representation Learning for Single-Source Domain Generalization in LiDAR Semantic Segmentation.

IEEE transactions on pattern analysis and machine intelligence·2026
Same author

Analysis of Multiscale Condensation Phenomena Using a Zero-Shot Computer Vision Framework.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same author

Temporal Stereo Matching From Event Cameras via Joint Learning With Stereoscopic Flow.

IEEE transactions on pattern analysis and machine intelligence·2025
Same author

Correction: Deep-learning-based gas identification by time-variant illumination of a single micro-LED-embedded gas sensor.

Light, science & applications·2025
Same author

Neuromorphic olfaction with ultralow-power gas sensors and ovonic threshold switch.

Science advances·2025
Same author

Multigas Identification by Temperature-Modulated Operation of a Single Anodic Aluminum Oxide Gas Sensor Platform and Deep Learning Algorithm.

ACS sensors·2025

Area of Science:

  • Computer Vision
  • Image Processing
  • Machine Learning

Background:

  • Correspondence search is crucial for tasks like 3D reconstruction.
  • Existing local methods often struggle with image ambiguity.
  • Improving accuracy in stereo matching remains an active research area.

Purpose of the Study:

  • To develop a more robust window-based method for correspondence search.
  • To reduce image ambiguity by adaptively weighting pixels within a support window.
  • To enhance the performance of local stereo matching techniques.

Main Methods:

  • A novel window-based approach for correspondence search.
  • Utilizing varying support-weights for pixels within a defined window.
  • Adjusting weights based on color similarity and geometric proximity to reduce image ambiguity.

Related Experiment Videos

Main Results:

  • The proposed method demonstrates superior performance compared to existing local methods.
  • Achieved state-of-the-art results on standard stereo benchmarks.
  • Effective reduction of image ambiguity through adaptive weighting.

Conclusions:

  • The new window-based method offers a significant improvement for correspondence search.
  • Adaptive support-weighting effectively addresses image ambiguity in stereo matching.
  • The approach shows strong potential for various computer vision applications requiring accurate image correspondence.