Will you publish the DINO based model that achieved SOTA on COCO?

Question

ronslos opened this issue 2 years ago · 7 comments

Hi, thanks for the interesting and insightful work!
Will you publish code and weights for the DINO based model that achieves SOTA on COCO detection?

Answer 1 · 2022-11-08T17:43:57.000Z

thanks for your interest, yes, we will release the large, xlarge and huge models shortly.

Answer 2 · 2022-11-13T09:45:06.000Z

HI, @ronslos , we just released the large, xlarge and huge ckpts pretrained on imagenet-22k. Give it a try!

Answer 3 · 2022-11-15T00:57:57.000Z

@jwyang , could you please release them for object detection?

Answer 4 · 2022-11-15T01:39:27.000Z

Hi, @YChunag , I will release the od pretrained ckpts at some points soon.

Answer 5 · 2022-11-15T02:32:40.000Z

Please refer to this repo for the object detection with our FocalNet: https://github.com/FocalNet/FocalNet-DINO

Answer 6 · 2022-11-30T07:45:04.000Z

Hello, Thank you for sharing your work.
I wanted to know if the above mentioned models for Object detection has been released.

Answer 7 · 2023-03-29T08:33:45.000Z

Hi, @jwyang!
Will you combine the FocalNet: https://github.com/FocalNet/FocalNet-DINO into this main repo?