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

Three-dimensional model-based object recognition and segmentation in cluttered scenes.

Ajmal S Mian1, Mohammed Bennamoun, Robyn Owens

  • 1School of Computer Science and Software Engineering, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. ajmal@csse.uwa.edu.au

IEEE Transactions on Pattern Analysis and Machine Intelligence
|September 22, 2006
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

Triple Spectral Fusion for Sensor-based Human Activity Recognition.

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

DynaPURLS: Dynamic Refinement of Part-Aware Representations for Skeleton-Based Zero-Shot Action Recognition.

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

Fully automated, deep learning, cardiac CT-based multimodal network for cardiovascular risk stratification in high-risk perioperative patients.

European heart journal. Digital health·2026
Same author

Trait Association for Flowering Time in Lentil from Global Multi-Environment Data Using GWAS and Machine Learning.

Plants (Basel, Switzerland)·2026
Same author

AquaticCLIP: A Vision-Language Foundation Model and Dataset for Underwater Scene Analysis.

IEEE transactions on neural networks and learning systems·2026
Same author

Unleashing the Power of Text-to-Image Diffusion Models for Category-Agnostic Pose Estimation.

IEEE transactions on pattern analysis and machine intelligence·2026
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

This study introduces a new 3D model-based algorithm for viewpoint-independent object recognition and segmentation. The method efficiently recognizes and segments free-form objects even with clutter and occlusions, achieving a 95% recognition rate.

Area of Science:

  • Computer Vision
  • Robotics
  • 3D Object Recognition

Background:

  • Viewpoint-independent recognition and segmentation of free-form objects in cluttered and occluded scenes remain significant challenges in computer vision.
  • Existing methods often struggle with efficiency and accuracy when dealing with complex real-world scenarios.

Purpose of the Study:

  • To present a novel, automatic, and efficient 3D model-based algorithm for viewpoint-independent object recognition and segmentation.
  • To demonstrate the algorithm's effectiveness in handling clutter and occlusions.

Main Methods:

  • Offline construction of 3D object models from unordered range images, converting views into multidimensional tensor representations.
  • Automatic establishment of correspondences between views using a hash table-based voting scheme on tensors, creating a graph for view registration.

Related Experiment Videos

  • Integration of registered views into seamless 3D models forming a model library.
  • Online recognition via simultaneous matching of scene tensors against the model library using voting, followed by similarity calculation and object segmentation.

    • Main Results:

    • Achieved an overall recognition rate of 95% on experiments with 55 models and 610 real and synthetic scenes.
    • Demonstrated superior recognition rate and efficiency compared to the spin images method.
    • Successfully segmented objects in scenes with clutter and occlusions.

    Conclusions:

    • The proposed 3D model-based algorithm offers an efficient and accurate solution for viewpoint-independent object recognition and segmentation.
    • The tensor representation and hash table-based voting scheme are effective for establishing correspondences and recognizing objects in complex scenes.
    • The algorithm shows significant advantages over existing methods like spin images in terms of performance.