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

Keypoint recognition using randomized trees.

Vincent Lepetit1, Pascal Fua

  • 1Ecole Polytechnique Fédérale de Lausanne, Computer Vision Laboratory, CH-1015 Lausanne, Switzerland. Vincent.Lepetit@epfl.ch

IEEE Transactions on Pattern Analysis and Machine Intelligence
|August 26, 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

Two Projections Suffice for Cerebral Vascular Reconstruction.

Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention·2025
Same author

Temporally-Consistent Surface Reconstruction Using Metrically-Consistent Atlases.

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

Harnessing deep learning to detect bronchiolitis obliterans syndrome from chest CT.

Communications medicine·2025
Same author

<i>MedShapeNet</i> - a large-scale dataset of 3D medical shapes for computer vision.

Biomedizinische Technik. Biomedical engineering·2024
Same author

Efficient anatomical labeling of pulmonary tree structures via deep point-graph representation-based implicit fields.

Medical image analysis·2024
Same author

Author Correction: BigNeuron: a resource to benchmark and predict performance of algorithms for automated tracing of neurons in light microscopy datasets.

Nature methods·2024
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 presents a fast keypoint-based method for 3D object detection and pose estimation. By shifting computation to training, the algorithm achieves real-time performance for various object types, including deformable ones.

Area of Science:

  • Computer Vision
  • Robotics
  • Machine Learning

Background:

  • Real-time 3D object detection and pose estimation are crucial for many applications.
  • Existing methods often face challenges with runtime performance, especially under varying conditions.

Purpose of the Study:

  • To develop a robust, accurate, and computationally efficient algorithm for 3D object detection and pose estimation.
  • To demonstrate the effectiveness of a keypoint-based approach with a focus on reducing runtime complexity.

Main Methods:

  • A keypoint-based approach was developed, treating wide-baseline matching as a classification problem.
  • Computational burden was shifted to a training phase, enabling the use of simpler keypoint detectors.
  • The system was trained using registered images of target objects.

Related Experiment Videos

Main Results:

  • The developed algorithm achieves frame-rate performance, offering significant reductions in runtime computational complexity.
  • A simple keypoint detector was shown to be sufficient for detection and tracking under large perspective and scale variations.
  • The system successfully detects planar, nonplanar, and deformable objects, and estimates poses and deformations.

Conclusions:

  • The proposed keypoint-based method provides a robust, accurate, and fast solution for 3D object detection and pose estimation.
  • Shifting computation to the training phase is an effective strategy for achieving real-time performance.
  • The approach is versatile, handling various object types and challenging viewing conditions.