Skip to main content

Research.

Computer vision research focused on interpretable models, anomaly detection, and annotation-free defect localization.

Simplified method diagram

    Fact-based diagram synthesized from the documented methodology; it is not an original paper figure.
    Active2025

    Prototype-Part Interpretable Image Recognition

    Research Assistant · Tulane University

    Explainable image classification using discovered object parts as interpretable prototypes.

    View research case study →

    Simplified method diagram

      Fact-based diagram synthesized from the documented methodology; it is not an original paper figure.
      Active2024

      Concept Distribution Modeling for Few-Shot Anomaly Detection

      Research Assistant · Tulane University

      Few-shot anomaly detection on industrial benchmarks without language prompts.

      96.3%MVTec AD pAUROC (1-shot)

      View research case study →

      Simplified method diagram

        Fact-based diagram synthesized from the documented methodology; it is not an original paper figure.
        Completed2023

        Unsupervised Crack Detection in X-ray Tomography using SAM

        Research Assistant · Tulane University

        Detects and delineates cracks in noisy tomographic images without manual annotation.

        View research case study →