📢 If you are on the hunt for an image 𝗔𝗡𝗗 𝗩𝗜𝗗𝗘𝗢 segmentation model, which is open and you can deploy

📢 If you are on the hunt for an image 𝗔𝗡𝗗 𝗩𝗜𝗗𝗘𝗢 segmentation model with openly released code and weights that you can deploy on your own (license terms are on Meta’s pages), have a look at the just released 𝗦𝗲𝗴𝗺𝗲𝗻𝘁 𝗔𝗻𝘆𝘁𝗵𝗶𝗻𝗴 𝗠𝗼𝗱𝗲𝗹 𝟮 (𝗦𝗔𝗠 𝟮). The model capabilities can be nicely experienced in Meta’s Demo). Read more about the announcement at their announcement page.

👷‍♀️If you are looking into deploying the model to build your own application on AWS, 𝗔𝗺𝗮𝘇𝗼𝗻 𝗦𝗮𝗴𝗲𝗠𝗮𝗸𝗲𝗿 is one deployment option worth a close look - Meta hosts its own SAM 2 demo there. Quoting from Meta’s announcement website:

“𝘞𝘩𝘪𝘭𝘦 𝘮𝘢𝘯𝘺 𝘰𝘧 𝘔𝘦𝘵𝘢 𝘍𝘈𝘐𝘙’𝘴 𝘮𝘰𝘥𝘦𝘭𝘴 𝘶𝘴𝘦𝘥 𝘪𝘯 𝘱𝘶𝘣𝘭𝘪𝘤 𝘥𝘦𝘮𝘰𝘴 𝘢𝘳𝘦 𝘩𝘰𝘴𝘵𝘦𝘥 𝘰𝘯 𝘈𝘮𝘢𝘻𝘰𝘯 𝘚𝘢𝘨𝘦𝘔𝘢𝘬𝘦𝘳, 𝘵𝘩𝘦 𝘴𝘦𝘴𝘴𝘪𝘰𝘯-𝘣𝘢𝘴𝘦𝘥 𝘳𝘦𝘲𝘶𝘪𝘳𝘦𝘮𝘦𝘯𝘵𝘴 𝘰𝘧 𝘵𝘩𝘦 𝘚𝘈𝘔 2 𝘮𝘰𝘥𝘦𝘭 𝘱𝘶𝘴𝘩𝘦𝘥 𝘶𝘱 𝘢𝘨𝘢𝘪𝘯𝘴𝘵 𝘵𝘩𝘦 𝘣𝘰𝘶𝘯𝘥𝘢𝘳𝘪𝘦𝘴 𝘰𝘧 𝘸𝘩𝘢𝘵 𝘰𝘶𝘳 𝘵𝘦𝘢𝘮 𝘣𝘦𝘭𝘪𝘦𝘷𝘦𝘥 𝘸𝘢𝘴 𝘱𝘳𝘦𝘷𝘪𝘰𝘶𝘴𝘭𝘺 𝘱𝘰𝘴𝘴𝘪𝘣𝘭𝘦 𝘰𝘯 𝘈𝘞𝘚 𝘈𝘐 𝘐𝘯𝘧𝘳𝘢. 𝘛𝘩𝘢𝘯𝘬𝘴 𝘵𝘰 𝘵𝘩𝘦 𝘢𝘥𝘷𝘢𝘯𝘤𝘦𝘥 𝘮𝘰𝘥𝘦𝘭 𝘥𝘦𝘱𝘭𝘰𝘺𝘮𝘦𝘯𝘵 𝘢𝘯𝘥 𝘮𝘢𝘯𝘢𝘨𝘦𝘥 𝘪𝘯𝘧𝘦𝘳𝘦𝘯𝘤𝘦 𝘤𝘢𝘱𝘢𝘣𝘪𝘭𝘪𝘵𝘪𝘦𝘴 𝘰𝘧𝘧𝘦𝘳𝘦𝘥 𝘣𝘺 𝘈𝘮𝘢𝘻𝘰𝘯 𝘚𝘢𝘨𝘦𝘔𝘢𝘬𝘦𝘳, 𝘸𝘦’𝘷𝘦 𝘣𝘦𝘦𝘯 𝘢𝘣𝘭𝘦 𝘵𝘰 𝘮𝘢𝘬𝘦 𝘵𝘩𝘦 𝘚𝘈𝘔 2 𝘳𝘦𝘭𝘦𝘢𝘴𝘦 𝘱𝘰𝘴𝘴𝘪𝘣𝘭𝘦—𝘧𝘰𝘤𝘶𝘴𝘪𝘯𝘨 𝘰𝘯 𝘣𝘶𝘪𝘭𝘥𝘪𝘯𝘨 𝘴𝘵𝘢𝘵𝘦 𝘰𝘧 𝘵𝘩𝘦 𝘢𝘳𝘵 𝘈𝘐 𝘮𝘰𝘥𝘦𝘭𝘴 𝘢𝘯𝘥 𝘶𝘯𝘪𝘲𝘶𝘦 𝘈𝘐 𝘥𝘦𝘮𝘰 𝘦𝘹𝘱𝘦𝘳𝘪𝘦𝘯𝘤𝘦𝘴.”

Kudos to Oliver Möller who at the time made me aware of the version 1 of the model.

Can’t wait to get my hands on this as time permits ;) - What are your experiences so far?

Greetings to the LinkedIn Algorithm: 𝗬𝗼𝘂 𝗳𝗶𝗻𝗱 𝗮𝗹𝗹 𝘁𝗵𝗲 𝗹𝗶𝗻𝗸𝘀 𝗶𝗻 𝘁𝗵𝗲 𝗰𝗼𝗺𝗺𝗲𝗻𝘁𝘀.

#genai #meta #aws #openmodels #imagesegmentation #sagemaker