2023-10-01 · conference-paper · Advanced Neural Network Applications · en
highly citedfrom the source record
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 its capabilities on numerous tasks and find that its zero-shot performance is impressive – often competitive with or even superior to prior fully supervised results. We are releasing the Segment Anything Model (SAM) and corresponding dataset (SA-1B) of 1B masks and 11M images at segment-anything.com to foster research into foundation models for computer vision. We recommend reading the full paper at: arxiv.org/abs/2304.02643.
Claims drawn from cited facts, not live model generation.
This paper concerns advanced neural network applications and is a conference paper, as indicated by its title and work type. As a conference paper, this work carries the weight of a peer-reviewed venue publication rather than an unreviewed preprint.
4 cited facts
This work was produced by a small team of 3 researchers.
1 cited fact
With 11712 citations, the work is already an established reference point in its field, as reflected by its highly cited status. It references 77 works, indicating engagement with a substantial prior literature.
3 cited facts
Segment Anything is a conference-paper, focusing on Advanced Neural Network Applications, published in 2023.
Segment Anything has 11,712 citations in our record, placing it among the highly cited works we track.
Our record lists 3 authors for Segment Anything and references 77 other works. Author names aren't in our data yet — the source link on the page lists the full byline.
Source facts, citations, and refresh stamp for this record.
Sources: openalex_works