CV4MS: Computer Vision for Materials Science

Overview

In materials science, a variety of microscopy image data (e.g., optical microscopy, electron microscopy, X-ray microscopy) are used to understand material structures and properties. Many of these microscopy techniques facilitate the rapid and cost-effective collection of large, complex image datasets that can overwhelm the available time of the subject matter experts required for interpretation. For example, large video and 3D datasets (e.g., produced by new high-speed electron detectors and X-ray computed tomography) are nearly impossible to manually quantify. 

Computer vision and machine learning are critical tools to support large-scale materials characterization and development of new materials. Quantified structure features that are extracted from the data can be leveraged in statistical and machine learning models that establish processing-structure-property-performance (PSPP) relationships to identify non-linear and unintuitive trends in the high dimensional materials development space further accelerating materials development. 

The aim of the Computer Vision for Materials Science (CV4MS) workshop is to bring together cross-disciplinary researchers to demonstrate recent advancements in machine learning, computer vision, and materials microscopy, and discuss open problems such as representation learning, uncertainty quantification, and explainability in materials microscopy analysis. There will be a focus on the unique challenges and advantages in materials characterization including limited ground-truth labels, multimodal datasets, and physics-based constraints, which could lead to new computer vision frameworks for the broader scientific imaging community.

This CV4MS@ICCV2025 workshop follows the CV4MS@CVPR2024 event held in conjunction with CVPR 2024, continuing our focus on advancing computer vision and machine learning for materials science.