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

Comparing object recognition from binary and bipolar edge features.

Jae-Hyun Jung1, Tian Pu1,2, Eli Peli1

  • 1Schepens Eye Research Institute, Massachusetts Eye and Ear, Harvard Medical School, Boston, MA.

IS&T International Symposium on Electronic Imaging
|April 15, 2017
PubMed
Summary
This summary is machine-generated.

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

Pedestrian Collision Detection and Avoidance in Cerebral Visual Impairment During Unrestricted Walking in an Immersive Virtual Reality Environment.

Research square·2026
Same author

A randomized crossover clinical trial of multi-periscopic and Fresnel peripheral prism glasses for hemianopia.

Optometry and vision science : official publication of the American Academy of Optometry·2026
Same author

Integrated sludge resources hierarchical recovery: enhancing VFAs production through pilot scale liquid-state anaerobic fermentation by sludge residues-based biochar.

Bioresource technology·2026
Same author

Correction: METTL3/YTHDF1 drives M1 macrophage polarization and aggravates ulcerative colitis progression by regulating m6A modification of SerpinB5 mRNA and FBXO32-dependent NF-κB pathway.

Journal of gastroenterology·2026
Same author

Effects of Sleep Duration on Electroencephalographic and Autonomic Nervous System Responses to High-Intensity Exercise.

Healthcare (Basel, Switzerland)·2026
Same author

Real-world safety profile and mechanistic insights into regorafenib-induced liver failure: a pharmacovigilance study integrated with network toxicology.

Frontiers in pharmacology·2026
Same journal

PySpatial: A High-Speed Whole Slide Image Pathomics Toolkit.

IS&T International Symposium on Electronic Imaging·2025
Same journal

Write Sentence with Images: Revisit the Large Vision Model with Visual Sentence.

IS&T International Symposium on Electronic Imaging·2025
Same journal

mTREE: Multi-Level Text-Guided Representation End-to-End Learning for Whole Slide Image Analysis.

IS&T International Symposium on Electronic Imaging·2025
Same journal

Scale-up Unlearnable Examples Learning with High-Performance Computing.

IS&T International Symposium on Electronic Imaging·2025
Same journal

Segment Anything Model (SAM) for Digital Pathology: Assess Zero-shot Segmentation on Whole Slide Imaging.

IS&T International Symposium on Electronic Imaging·2025
Same journal

Egocentric Boundaries on Distinguishing Colliding and Non-Colliding Pedestrians while Walking in a Virtual Environment.

IS&T International Symposium on Electronic Imaging·2024
See all related articles

Bipolar edges, which retain depth cues lost in binary edge images, significantly improve object recognition, especially in complex visual scenes. This enhances feature representation for better scene understanding.

Area of Science:

  • Computer Vision
  • Image Processing
  • Human Perception

Background:

  • Edges in images are crucial for object recognition.
  • Traditional binary edge representations lose depth information.
  • This lost depth information is related to shading cues.

Purpose of the Study:

  • To investigate the impact of bipolar edges on object recognition.
  • To determine if bipolar edges restore lost depth information.
  • To compare recognition rates using binary versus bipolar edge features.

Main Methods:

  • Presented 16 types of binary and bipolar edge features to 26 human subjects.
  • Recorded and analyzed object recognition rates for each feature type.
  • Evaluated performance across simple and complex background scenes.

Related Experiment Videos

Main Results:

  • Object recognition rates were consistently higher with bipolar edge features.
  • The improvement in recognition was statistically significant in scenes with complex backgrounds.
  • Bipolar edges demonstrated a superior ability to convey depth information.

Conclusions:

  • Bipolar edge representations offer advantages over binary edges for object recognition.
  • Restoring depth information via bipolar edges enhances scene understanding.
  • Bipolar features show promise for improving visual perception tasks.