Meta, the company that owns Facebook, has announced the release of its new artificial intelligence (AI) model called Segment Anything Model (SAM). According to the company’s research division, SAM can identify individual objects in images and videos, even if it has not been trained to do so. Additionally, Meta has published the largest dataset of image annotations ever of its kind.
Using SAM, objects can be selected by either clicking on them or writing text prompts. For example, typing the word “cat” prompts the tool to draw boxes around cats in a photo.
Meta has been working on developing generative AI, which creates new content instead of categorizing data. The company’s CEO, Mark Zuckerberg, has stated that incorporating such generative AI “creative aids” into Meta’s apps is a priority this year. Meta is already using technology similar to SAM for tagging photos, moderating prohibited content, and recommending posts to users of Facebook and Instagram. The company intends to make SAM and its dataset available for download under a non-commercial license. Users who upload their own images to an accompanying prototype must agree to use it only for research purposes.