Interactive Visual Feature Search

TL;DR

This paper introduces Visual Feature Search, an interactive visualization tool for CNNs that helps users interpret model behavior by searching for similar features across datasets, adaptable to various applications.

Contribution

The paper presents a novel, adaptable interactive visualization tool for CNNs that enhances interpretability and can be integrated into researchers' workflows.

Findings

Effective in medical imaging analysis

Useful for wildlife classification tasks

Facilitates understanding of model features

Abstract

Many visualization techniques have been created to explain the behavior of computer vision models, but they largely consist of static diagrams that convey limited information. Interactive visualizations allow users to more easily interpret a model's behavior, but most are not easily reusable for new models. We introduce Visual Feature Search, a novel interactive visualization that is adaptable to any CNN and can easily be incorporated into a researcher's workflow. Our tool allows a user to highlight an image region and search for images from a given dataset with the most similar model features. We demonstrate how our tool elucidates different aspects of model behavior by performing experiments on a range of applications, such as in medical imaging and wildlife classification.