AI Image Segmentation
Click on any object to cut it out. Uses Meta's Segment Anything AI. Runs entirely in your browser — your images never leave your device.
Processed on your device — never uploaded
Upload an Image
Your image never leaves your deviceWorks offline after first use
Frequently Asked Questions
What is AI image segmentation?
AI image segmentation identifies and separates individual objects in an image. This tool uses Meta's Segment Anything AI (SlimSAM) to let you click on any object and get a precise cutout with a transparent background, perfect for photo editing, design, and compositing.
How do the include and exclude clicks work?
Left-click (green dot) marks a point as part of the object you want to select. Right-click (red dot) marks a point as something to exclude from the selection. This helps refine the segmentation — for example, click on a person to select them, then right-click on something behind them that was accidentally included.
Is my image uploaded to a server?
No. All processing happens entirely on your device, in your browser. The AI engine (~14 MB) is downloaded once and cached. Your images never leave your device, ensuring complete privacy.
Why is the first use slower?
On first use, the AI engine (~14 MB) needs to be downloaded to your browser. This is cached automatically, so subsequent uses are instant. After loading, the AI analyzes your image once, then each click generates a cutout in under a second.