Reproduce “MaskTerial: A Foundation Model for Automated 2D Material Flake Detection” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Jan-Lucas Uslu, Alexey Nekrasov, Alexander Hermans, Bernd Beschoten, Bastian Leibe, Lutz Waldecker, Christoph Stampfer

MaskTerial finds exfoliated flakes in microscope images and separately classifies layer thickness. It pretrains instance segmentation on simulated images made from unlabeled real flake shapes, then adapts using a few real images. Across eight datasets, synthetic pretraining raises AP50 with a GMM head from 2.7 ± 2.8 to 66.8 ± 10.6 at 10 images per class, within this evaluation.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.