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

Learning Topology-Aware Representations via Test-Time Adaptation for Anomaly Segmentation.

Ali Zia, Usman Ali, Abdul Rehman

    IEEE Transactions on Pattern Analysis and Machine Intelligence
    |July 3, 2026
    PubMed
    Summary

    TopoTTA enhances anomaly segmentation by using topological data analysis to maintain structural consistency during test-time adaptation (TTA). This novel framework significantly improves detection of complex defects without retraining models.

    Related Concept Videos

    Survival Tree01:19

    Survival Tree

    Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
     Building a Survival Tree
    Constructing a survival tree begins...

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Development and preliminary validation of the Meat Consumption Scale.

    BMC psychology·2025
    Same author

    Safety and efficacy of mepolizumab in eosinophilic chronic obstructive pulmonary disease: a systematic review and meta-analysis.

    Expert review of respiratory medicine·2025
    Same author

    Cancer Pain Is Not One-Size-Fits-All: Evolving from Tradition to Precision.

    Clinics and practice·2025
    Same author

    The role of mineralization bacteria in the immobilization of cadmium and lead in aqueous solutions.

    Journal of hazardous materials·2025
    Same author

    Occult Femoral Neck Fracture Misdiagnosed as Septic Arthritis: A Case Highlighting Diagnostic Challenges in Busy Emergency Settings.

    Cureus·2025
    Same author

    Comment on "Misaligned Attitudes and Perceptions Among Adolescents Living With Obesity, Caregivers and Healthcare Professionals: ACTION Teens Australia Survey Study".

    Journal of paediatrics and child health·2025

    Area of Science:

    • Computer Vision
    • Machine Learning
    • Topological Data Analysis

    Background:

    • Test-time adaptation (TTA) is crucial for deep learning models facing distribution shifts.
    • Current TTA for anomaly segmentation struggles with structural consistency due to pixel-level heuristics.
    • Existing methods often ignore complex spatial relationships in defect geometries.

    Purpose of the Study:

    • To introduce TopoTTA, a novel framework integrating persistent homology into TTA for anomaly segmentation.
    • To enforce geometric and structural coherence during adaptation without retraining.
    • To improve segmentation quality, especially for anomalies with complex shapes.

    Main Methods:

    • TopoTTA utilizes persistent homology from topological data analysis within the TTA pipeline.

    Related Experiment Videos

  • Multi-level cubical complex filtration is applied to anomaly score maps.
  • Robust topological pseudo-labels guide a lightweight test-time classifier.

    • Main Results:

    • TopoTTA achieves an average 15% F1 improvement over state-of-the-art methods across six benchmarks.
    • Significant gains are observed on anomalies with complex geometric or structural variations.
    • The method preserves connectivity and generalizes across 2D and 3D modalities.

    Conclusions:

    • Integrating topological reasoning into TTA offers a principled approach to structure-aware generalization.
    • TopoTTA bridges the gap between geometric learning and robust adaptation in anomaly segmentation.
    • The framework enhances segmentation quality by preserving structural integrity.