Meta shares code for a new AI model that can \”cut out any object\” within an image.

Reuters reports that Meta announced on Wednesday an AI model named the Segment Any Model (SAM), which can identify objects in videos and images, even if they were not seen during training.

According to a Meta blog post, SAM is a model for image segmentation that can be activated by text or clicks from the user to isolate certain objects in an image. Image segmentation in computer vision is the process of dividing an image up into segments or regions that each represent a specific area or object.

Source:
https://arstechnica.com/information-technology/2023/04/meta-introduces-ai-model-that-can-isolate-and-mask-objects-within-images/?utm_medium=social&utm_source=facebook&utm_brand=ars&utm_social-type=owned