Tag: Segment Anything Model

In a prompted evaluation of three Aortic Vessel Tree cohorts and one Type-B Aortic Dissection dataset, VesselSAM achieved slice-averaged Dice ...

Plant pathologists have long faced a stubborn diagnostic problem: a single leaf can carry several diseases at once, each with ...

Researchers have adapted Meta's Segment Anything Model for medical imaging by combining low-rank fine-tuning, Mamba-based adapters, and a prompt-free decoder, ...

Researchers in Serbia have built and validated a modular collaborative robotic workstation that combines foundation-model perception with safe human-robot collaboration ...

Researchers at the University of Potsdam have developed OrthoSAM, an AI workflow that automatically delineates thousands of individual river pebbles ...

Researchers have retrained a biomedical AI model to automatically measure sediment grains with record accuracy across photographs and X-ray scans, ...

Researchers combined aerial image segmentation and camera-based object detection to build the first corridor-wide, AI-driven indicators of parking supply and ...

A Spanish study of 126 primary school pupils found that combining structured hand hygiene education with UV visual feedback and ...

Researchers at the University of Bologna and University of Florence have developed a human-AI co-creative pipeline that generates evocative eco-visualizations ...

Researchers in Toronto fine-tuned the Segment Anything Model to automatically segment footwear outsoles, using the resulting contact areas to predict ...