arXiv 2304.02643

Segment Anything

By Alexander Kirillov, Eric Mintun, et al.

Published 2023-04-05

Wiki summary

Explore the paper's summary, context, and related research on Papiers.

We introduce the Segment Anything (SA) project: a new task, model, and dataset for image segmentation. Using our efficient model in a data collection loop, we built the largest segmentation dataset to date (by far), with over 1 billion masks on 11M licensed and privacy respecting images. The model is designed and trained to be promptable, so it can transfer zero-shot to new image distributions and tasks. We evaluate…

View the original paper on arXiv