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

Motion segmentation using occlusions.

Abhijit S Ogale1, Cornelia Fermüller, Yiannis Aloimonos

  • 1Center for Automation Research, Department of Computer Science, University of Maryland, College Park, MD 20742, USA. ogale@cfar.umd.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|June 10, 2005
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

Single-Step Latent Diffusion for Underwater Image Restoration.

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

Minimal perception: enabling autonomy in resource-constrained robots.

Frontiers in robotics and AI·2024
Same author

Exploring synthetic datasets for computer-aided detection: a case study using phantom scan data for enhanced lung nodule false positive reduction.

Journal of medical imaging (Bellingham, Wash.)·2024
Same author

Microsaccade-inspired event camera for robotics.

Science robotics·2024
Same author

Editorial: Enhanced human modeling in robotics for socially-aware place navigation.

Frontiers in robotics and AI·2024
Same author

Motor-invariants for action understanding in video.

Physics of life reviews·2023
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

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

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

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

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

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

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

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

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

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

Learning Shape Anchors for Holistic Indoor Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

Occlusions are crucial for identifying independently moving objects in videos from moving cameras. This study introduces a novel algorithm to fill occlusions and determine ordinal depth, enhancing object detection capabilities.

Area of Science:

  • Computer Vision
  • Robotics
  • Artificial Intelligence

Background:

  • Analyzing video motion from moving cameras is complex due to combined camera motion, scene structure, and object movement.
  • Restricted fields of view and noise can lead to ambiguous 3D motion estimates.
  • Existing methods may struggle with precise object motion detection in dynamic scenes.

Purpose of the Study:

  • To investigate the critical role of occlusions in detecting independently moving objects.
  • To develop a method for inferring ordinal depth using occlusion information.
  • To introduce a novel algorithm for occlusion filling and depth estimation.

Main Methods:

  • Utilizing optical flow to analyze motion fields.
  • Employing motion-based clustering for object segmentation.

Related Experiment Videos

  • Developing a novel algorithm for occlusion filling and ordinal depth deduction.
  • Comparing structure from motion estimates with external sources (e.g., stereo vision).

    • Main Results:

    • Occlusion information is vital for identifying independently moving objects.
    • A novel algorithm effectively fills occlusions and deduces ordinal depth.
    • The method enables the detection of a new class of moving objects.
    • Successful integration of occlusion data enhances motion analysis.

    Conclusions:

    • Occlusion handling is a key factor in robustly detecting independently moving objects.
    • The proposed algorithm provides a significant advancement in inferring ordinal depth from video.
    • This research contributes to improved scene understanding and object tracking in dynamic environments.