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

Robust photometric invariant features from the color tensor.

Joost van de Weijer1, Theo Gevers, Arnold W M Smeulders

  • 1Intelligent Sensory Information Systems, University of Amsterdam, Kruislaan 403, 1098 SJ Amsterdam, The Netherlands. joostw@science.uva.nl

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|January 27, 2006
PubMed
Summary

This study introduces robust photometric invariant features by integrating color tensor analysis with photometric invariance theory. This approach enhances feature detection in computer vision, improving robustness to various scene conditions and photometric changes.

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

Deepfake perpetrator conversation for adults with sexual abuse-related posttraumatic stress disorder: intervention development and multiple baseline study protocol.

European journal of psychotraumatology·2026
Same author

Virtual rescripting after loss using deepfake technology in prolonged grief treatment: a study protocol for a multiple baseline design.

European journal of psychotraumatology·2026
Same author

Trust Your Good Friends: Source-Free Domain Adaptation by Reciprocal Neighborhood Clustering.

IEEE transactions on pattern analysis and machine intelligence·2023
Same author

Generative Multi-Label Zero-Shot Learning.

IEEE transactions on pattern analysis and machine intelligence·2023
Same author

Interactive Learning of Intrinsic and Extrinsic Properties for All-Day Semantic Segmentation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2023
Same author

Geometric Back-Propagation in Morphological Neural Networks.

IEEE transactions on pattern analysis and machine intelligence·2023

Area of Science:

  • Computer Vision
  • Image Processing
  • Photometry

Background:

  • Luminance-based features are standard in computer vision, but can lose information with isoluminant color data.
  • Exploiting color data requires handling its vector nature and combining feature detection with photometric invariance theory.

Purpose of the Study:

  • To develop a framework for robust photometric invariant feature detection that fully utilizes color information.
  • To address the instability of traditional photometric invariants.

Main Methods:

  • Utilized the color tensor to manage the vector nature of color images.
  • Combined color tensor features with photometric invariant derivatives.
  • Derived an uncertainty measure for photometric invariant derivatives and incorporated it into the color tensor.

Related Experiment Videos

Main Results:

  • Developed robust photometric invariant features by incorporating an uncertainty measure.
  • Demonstrated the detection of features like edges, corners, optical flow, and curvature.
  • Experimental results confirm robustness to incidental events and improved applicability of invariants.

Conclusions:

  • The proposed method effectively combines photometric invariance theory and tensor-based features for robust image analysis.
  • The uncertainty measure enhances the reliability and applicability of photometric invariant features in computer vision.
  • This approach offers a significant advancement in exploiting color information for feature detection.