[R] Boundary-Metric Evaluation for Thin-Structure Segmentation under 2% Foreground Sparsity
Reddit - Machine Learning
1 min read
About this article
Hey! I'm currently a undergrad student graduating in May and soon starting my Masters in AI. I've wanted to write a research paper to start gaining some experience in that area and just recently finished my first one. This paper focuses on investigating segmentation under some extreme foreground sparsity, around 1.8% of positive pixels during a whiteboard digitization. It connects to a small project I was working on where you can take a photo of a whiteboard and it would identify what is actu...
You've been blocked by network security.To continue, log in to your Reddit account or use your developer tokenIf you think you've been blocked by mistake, file a ticket below and we'll look into it.Log in File a ticket
Originally published on March 03, 2026.
Curated by AI News.
![[2506.22504] Patch2Loc: Learning to Localize Patches for Unsupervised Brain Lesion Detection](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}