yatengLG/ISAT_with_segment_anything

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

2166 stars · 212 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

ISAT_with_segment_anything is an interactive semi-automatic image annotation tool built on the Segment Anything Model family (SAM, SAM2, SAM3, sam-hq, MobileSAM, EdgeSAM). It supports point, box, text, and visual prompts for fast segmentation labeling, plus a plugin system for extending functionality.

Use cases

• annotate images for semantic segmentation datasets
• create segmentation masks with SAM click prompts
• label objects in images using text prompts
• annotate video frames for segmentation
• export masks for training computer vision models
• auto-annotate images with YOLO models

When to choose

• you need fast semi-automatic segmentation labels for training data
• you want a GUI annotation tool leveraging SAM-family models
• you need text or visual prompt based labeling with SAM3
• you want an extensible annotation tool with plugins

When to avoid

• you only need bounding-box annotation for detection tasks
• you need a web-based collaborative annotation platform for teams
• you lack a GPU or the resources to run SAM models locally

Facets

application · maturity active

image-processing computer-vision machine-learning gui plugin-system computer-vision image-processing machine-learning artificial-intelligence developer-tools python cross-platform segment-anything image-annotation labeling-tool sam sam2 sam3 semantic-segmentation video-segmentation semi-automatic-annotation desktop

Member repositories

For agents

markdown · JSON · MCP: product_card(name="yatengLG/ISAT_with_segment_anything")