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

Fast two-frame multiscale dense optical flow estimation using discrete wavelet filters.

Haiying Liu1, Rama Chellappa, Azriel Rosenfeld

  • 1Center for Automation Research, University of Maryland, College Park, Maryland 20770, USA. hyliu@cfar.umd.edu

Journal of the Optical Society of America. A, Optics, Image Science, and Vision
|August 27, 2003
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

Effect and mechanism of Lycium barbarum polysaccharide on gastrointestinal motility in slow transit constipation.

Naunyn-Schmiedeberg's archives of pharmacology·2024
Same author

A clinical prognostic model related to T cells based on machine learning for predicting the prognosis and immune response of ovarian cancer.

Heliyon·2024
Same author

Efficacy and Safety of C3 Laminectomy Combined with Open-Door Laminoplasty versus Open-Door Laminoplasty Alone: A Systematic Review and Meta-Analysis.

World neurosurgery·2024
Same author

Optimized personalized management approach for moderate/severe OHSS: development and prospective validation of an OHSS risk assessment index.

Human reproduction (Oxford, England)·2024
Same author

Effect of single blastocyst-stage versus single cleavage-stage embryo transfer on cumulative live births in women with good prognosis undergoing in vitro fertilization: Multicenter Randomized Controlled Trial.

Nature communications·2024
Same author

NSUN2 is critical for mouse oocyte meiosis and early embryonic development.

Reproduction (Cambridge, England)·2024
Same journal

Multi-module collaborative optimization-driven fast speckle correlation imaging in variable environments.

Journal of the Optical Society of America. A, Optics, image science, and vision·2026
Same journal

Secrecy performance analysis of NOMA-UWOC systems over a vertically stratified WGG oceanic turbulence channel.

Journal of the Optical Society of America. A, Optics, image science, and vision·2026
Same journal

Backscattering of plane waves in a composite system containing a rough surface and anisotropic scatterers.

Journal of the Optical Society of America. A, Optics, image science, and vision·2026
Same journal

Aspherical surface construction methods based on extended Jacobi polynomials.

Journal of the Optical Society of America. A, Optics, image science, and vision·2026
Same journal

OCT sidelobe suppression method based on dual-path phase sinusoidal modulation and minimum value fusion.

Journal of the Optical Society of America. A, Optics, image science, and vision·2026
Same journal

Optical design concepts using wavelength-selective diffractive optics to enable miniaturized multimodal endoscopic imaging across separated spectral ranges.

Journal of the Optical Society of America. A, Optics, image science, and vision·2026
See all related articles

A novel multiscale algorithm estimates dense optical flow using wavelet filters, improving accuracy and overcoming limitations of traditional pyramid methods. This efficient O(N) approach offers robust performance for image sequence analysis.

Area of Science:

  • Computer Vision
  • Image Processing
  • Signal Processing

Background:

  • Dense optical flow estimation is crucial for analyzing motion in image sequences.
  • Traditional pyramid methods for optical flow suffer from error propagation and the 'flattening-out' problem in low-texture regions.
  • Wavelet decomposition offers a hierarchical image representation suitable for multiscale analysis.

Purpose of the Study:

  • To propose a novel multiscale algorithm for accurate and efficient dense optical flow estimation.
  • To overcome the limitations of traditional coarse-to-fine pyramid methods.
  • To develop an algorithm with a linear complexity of O(N) for practical applications.

Main Methods:

  • A multiscale framework integrating wavelet decomposition with differential techniques.

Related Experiment Videos

  • Utilizing specific compactly supported wavelet bases with one vanishing moment for hierarchical image, derivative, and corner representations.
  • Employing three components of wavelet decomposition and a 2D affine motion model to formulate optical flow as a linear system.
  • Solving the linear system simultaneously across all resolutions (coarse-and-fine).

    • Main Results:

    • The proposed algorithm achieves O(N) complexity, where N is the number of pixels.
    • It effectively addresses the 'flattening-out' problem encountered in traditional methods.
    • Simultaneous resolution processing improves accuracy and simplifies hardware implementation compared to coarse-to-fine approaches.
    • Experimental results demonstrate the algorithm's effectiveness and robustness across various image sequences.

    Conclusions:

    • The developed multiscale wavelet-based algorithm provides an accurate, efficient, and robust solution for dense optical flow estimation.
    • This method offers significant advantages over traditional pyramid techniques, particularly in handling low-texture areas.
    • The algorithm's linear complexity and simplified implementation make it suitable for real-time applications and hardware deployment.