Interactive Classification for Deep Learning Interpretation

TL;DR

This paper introduces an interactive web-based system that allows users to manipulate images and observe how deep learning classifiers respond, revealing insights into model robustness and feature reliance in real time.

Contribution

The authors develop an in-browser interactive tool for exploring deep learning image classification, enabling real-time image manipulation and analysis of model sensitivity.

Findings

Revealed surprising model failures and resilience through interactive image modifications.

Enabled real-time comparison of human and machine focus regions in images.

Demonstrated the system at CVPR, showcasing practical insights into model behavior.

Abstract

We present an interactive system enabling users to manipulate images to explore the robustness and sensitivity of deep learning image classifiers. Using modern web technologies to run in-browser inference, users can remove image features using inpainting algorithms and obtain new classifications in real time, which allows them to ask a variety of "what if" questions by experimentally modifying images and seeing how the model reacts. Our system allows users to compare and contrast what image regions humans and machine learning models use for classification, revealing a wide range of surprising results ranging from spectacular failures (e.g., a "water bottle" image becomes a "concert" when removing a person) to impressive resilience (e.g., a "baseball player" image remains correctly classified even without a glove or base). We demonstrate our system at The 2018 Conference on Computer Vision and Pattern Recognition (CVPR) for the audience to try it live. Our system is open-sourced at https://github.com/poloclub/interactive-classification. A video demo is available at https://youtu.be/llub5GcOF6w.