This week, grounding DINO 1.5 was released It is a new model that uses text prompts to detect objects from videos and images in real-time Examples & demo to try below:

1) Video object detection

2) Image object detection

3) Paper: Website: Playground:

4) Grounding DINO 1.5

Someone, please combine this with Segment Anything and replace the boxes with masks of the objects. It would be incredible!

I agree! That would be super helpful for editing

This would also rule out copyright violations because the law doesn’t distinguish between human and bot/code “seeing” and learning. What a great loop hole that was prolly not intended but awesome haha 😆

I never even considered this angle, Dustin! 👀😯I wonder if that is going to be an argument that a lot of these companies that have been asking for permission to train on user phone and car data will use. It is allowing their AI to see the world without scraping

wow

Agree!