Plans

Segmentation

Upload an image to get pixel-precise instance masks.

Requests remaining this month: 100 / 100

How it works

Upload any image and the model will outline the objects it recognises with pixel-precise masks. The base model is pretrained on the COCO dataset and covers 80 object classes.

0.200.85

Drop an image here

or click to browse

JPG, PNG, WebP -- max 10 MB

Or try a sample