On April 5th, Meta – the parent company of Facebook – introduced an artificial intelligence (AI) model capable of recognizing every detail in an image.
Meta logo on a phone screen in Moscow, Russia. (Photo: AFP/TTXVN).
In a blog post, Meta’s research department announced that the Segment Anything Model (SAM) can identify objects in images and videos even if it has never been trained on those particular objects. When using SAM, users can select objects in a photo by clicking on them or typing a prompt. For example, simply typing the phrase “cat” will prompt the AI to immediately draw a bounding box around each cat in the image.
Meta launched this new AI model amid a competitive landscape where many tech giants are racing to dominate this field. The company that owns Facebook is also internally testing several features based on Generative AI—an AI focused on creating entirely new content, rather than just recognizing or classifying data as other AIs do. A popular Generative AI tool today is ChatGPT.
Meta CEO Mark Zuckerberg stated that integrating AI-powered creative tools into Meta’s applications is the company’s top priority for 2023. So far, Meta has utilized technology similar to SAM internally for various functions such as photo tagging, content moderation, and determining recommended content for users on Facebook and Instagram.
Meta revealed that the launch of SAM will broaden access to this type of technology. Users will be able to download the SAM model and dataset for non-commercial use.
