DINOv3
github.com
DINOv3
1–10 of 32 posts
Re: DINOv3
#2Re: DINOv3
#3I have no idea what this even is.
Re: DINOv3
#4I have no idea what this even is.
D3NO?
Re: DINOv3
#5I have no idea what this even is.
> An extended family of versatile vision foundation models producing high-quality dense features and achieving outstanding performance on various vision tasks including outperforming the specialized state of the art across a broad range of settings, without fine-tuning
Re: DINOv3
#6I have no idea what this even is.
> An extended family of versatile vision foundation models producing high-quality dense features and achieving outstanding performance on various vision tasks including outperforming the specialized state of the art across a broad range of settings, without fine-tuning
English, doc
Re: DINOv3
#7I have no idea what this even is.
> An extended family of versatile vision foundation models producing high-quality dense features and achieving outstanding performance on various vision tasks including outperforming the specialized state of the art across a broad range of settings, without fine-tuning
To elaborate, this is a foundation model. This basically means it can take an arbitrary image and map it to a high dimensional space H in which ~arbitrary characteristics become much easier to solve for.
For example (and this might be oversimplifying a bit, computer vision people please correct me if I’m wrong) if you’re interested in knowing whether or not the image contains a cat, then maybe there is some hyperplane P in H for which images on one side of P do not contain a cat, and images on the other side do contain a cat. And so solving for “Does this image contain a cat?”becomes a much easier problem, all you have to do is figure out what P is. Once you do that, you can pass your image into DINO, dot product with the equation for P, and check whether the answer is negative or positive. The point is that finding P is much easier than training your own computer vision model from scratch.
Re: DINOv3
#8That's awesome. DINOv2 was the best image embedder until now.
Re: DINOv3
#9Earlier quoted context omitted.
> An extended family of versatile vision foundation models producing high-quality dense features and achieving outstanding performance on various vision tasks including outperforming the specialized state of the art across a broad range of settings, without fine-tuning
To elaborate, this is a foundation model. This basically means it can take an arbitrary image and map it to a high dimensional space H in which ~arbitrary characteristics become much easier to solve for. For example (and this might be oversimplifying a bit, computer vision people please correct me if I’m wrong) if you’re interested in knowing whether or not the image contains a cat, then maybe there is some hyperplan…
If computer vision were semantic search, nailed it. It’s a little more complicated than that but - with this new model, not by much :D
Re: DINOv3
#10You have to share your contact information, including DoB, and then be approved access, to obtain the models, and given that it's Meta I assume they're actually validating it against their All Humans database.
They made their own DINOv3 license for this release (whereas DINOv2 used the Apache 2.0 license).
Neat though. Will still check it out.
As a first comment, I had to install the latest transformer==4.56.0dev (e.g. pip install git+https://github.com/huggingface/transformers) for it to work properly. 4.55.2 and earlier was failing with a missing image type in the config.