Visual Understanding of Multiple Attributes Learning Model of X-Ray Scattering Images

TL;DR

This paper introduces a visualization system that helps domain scientists understand deep learning models extracting multiple attributes from X-ray scattering images by exploring feature space, classification outputs, and label correlations.

Contribution

It presents a novel visualization tool tailored for analyzing deep learning models on X-ray scattering data, focusing on multiple structural attributes and interactive exploration.

Findings

System enables detailed visual analysis of model behavior

Case studies demonstrate its effectiveness and usefulness

Facilitates understanding of attribute-related model decisions

Abstract

This extended abstract presents a visualization system, which is designed for domain scientists to visually understand their deep learning model of extracting multiple attributes in x-ray scattering images. The system focuses on studying the model behaviors related to multiple structural attributes. It allows users to explore the images in the feature space, the classification output of different attributes, with respect to the actual attributes labelled by domain scientists. Abundant interactions allow users to flexibly select instance images, their clusters, and compare them visually in details. Two preliminary case studies demonstrate its functionalities and usefulness.