SAM Cell Segmentation

소개

Provides expert guidance for building deep learning inference pipelines focused on biological cell segmentation using the Segment Anything Model (SAM) and MobileSAM. It streamlines the process of converting complex binary masks into precise polygon coordinates, ensuring data type consistency for downstream analysis. By offering standardized patterns for CLI tool design, SAM integration, and OpenCV-based mask processing, this skill helps developers create robust tools for processing microscopy imagery and CSV-based coordinate data.

주요 기능

• MobileSAM and SAM model integration patterns
• Optimized binary mask-to-polygon conversion using OpenCV
• 16 GitHub stars
• Automated verification checklists for interface and output validation
• Guidance on maintaining data type consistency for ML pipelines
• Standardized CLI design for deep learning inference scripts

사용 사례

• Processing biological microscopy images to extract cell boundaries
• Building command-line tools for automated object segmentation tasks
• Converting deep learning mask outputs into portable CSV/JSON coordinate formats