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

Drift detection and removal for sequential structure from motion algorithms.

Kurt Cornelis1, Frank Verbiest, Luc Van Gool

  • 1Katholieke Universiteit Leuven, Kasteelpark Arenberg 10, B-3001 Leuven-Heverlee, Belgium. kurt.cornelis@esat.kuleuven.ac.be

IEEE Transactions on Pattern Analysis and Machine Intelligence
|January 12, 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

Learning From Each Other: Generalized Federated Incremental Semantic Segmentation.

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

ACDC: The Adverse Conditions Dataset With Correspondences for Robust Semantic Driving Scene Perception.

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

UniDepthV2: Universal Monocular Metric Depth Estimation Made Simpler.

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

Subgrapher: visual fingerprinting of chemical structures.

Journal of cheminformatics·2025
Same author

Spatial-Temporal Graph Mamba for Music-Guided Dance Video Synthesis.

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

DiffI2I: Efficient Diffusion Model for Image-to-Image Translation.

IEEE transactions on pattern analysis and machine intelligence·2025
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

Drift in Structure from Motion (SfM) causes errors in 3D reconstructions. This study introduces drift detection and removal methods to improve SfM accuracy in video sequences.

Area of Science:

  • Computer Vision
  • Robotics
  • Photogrammetry

Background:

  • Sequential Structure from Motion (SfM) algorithms accumulate drift errors in extended image or video sequences.
  • Drift causes feature tracks to yield distinct 3D reconstructions for the same scene point, hindering accurate global optimization.
  • Bundle adjustment, a nonlinear optimization technique, requires drift removal from initial solutions for convergence to the true global optimum.

Purpose of the Study:

  • To develop and validate a method for detecting and removing drift in sequential Structure from Motion.
  • To enable accurate 3D reconstructions from long image or video sequences by mitigating cumulative errors.

Main Methods:

  • Drift detection by identifying 3D reconstructions of the same scene part differing only by a projective transformation.

Related Experiment Videos

  • Removal of drift from future processed images.
  • Adapted Bundle Adjustment utilizing correspondences from drift detection to correct previous images.

    • Main Results:

    • Drift detection effectively identifies projective transformations in reconstructed scene parts.
    • Drift removal significantly improves the accuracy of 3D reconstructions in sequential SfM.
    • Experiments on real video sequences confirm the effectiveness of the proposed drift detection and removal approach.

    Conclusions:

    • Drift detection is crucial for accurate Structure from Motion in long sequences.
    • The proposed Adapted Bundle Adjustment effectively corrects drift errors, leading to more reliable 3D reconstructions.
    • This work advances SfM by providing a robust solution to the persistent problem of drift accumulation.