Saved in:
Bibliographic Details
Main Authors: Lu, Mingjian, Tripathi, Pawan K., Shteyn, Mark, Ganguly, Debargha, French, Roger H., Chaudhary, Vipin, Wu, Yinghui
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.04154
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Segmentation architectures are typically benchmarked on single imaging modalities, obscuring deployment-relevant performance variations: an architecture optimal for one modality may underperform on another. We present a cross-modal evaluation framework for materials image segmentation spanning SEM, AFM, XCT, and optical microscopy. Our evaluation of six encoder-decoder combinations across seven datasets reveals that optimal architectures vary systematically by context: UNet excels for high-contrast 2D imaging while DeepLabv3+ is preferred for the hardest cases. The framework also provides deployment feedback via out-of-distribution detection and counterfactual explanations that reveal which microstructural features drive predictions. Together, the architecture guidance, reliability signals, and interpretability tools address a practical gap in materials characterization, where researchers lack tools to select architectures for their specific imaging setup or assess when models can be trusted on new samples.