Web version support for labeling

Thanks for the fantastic research and the code. It is very beneficial as it supports zero shot segmentation

Can we use the pre-trained models for custom images?
Researchers can also use this for labeling their images if we can use the [web tool](https://segment-anything.com/demo) and perform labeling and create a JSON/COCO format file.
Any ideas?

Start by reviewing the linked web tool and the repository's example notebooks to understand how pre-trained models are used with custom images. Done would mean a web-based labeling flow that supports custom images and produces JSON or COCO-format output.

Written by the indexing model from the issue text.