Quyans
Release DYG model on Hugging Face
Hi @Quyans 🤗
I'm Niels from the Hugging Face open-source team. I came across your impressive work on arXiv ("Drag Your Gaussian") and its accompanying project page (https://quyans.github.io/Drag-Your-Gaussian/). The videos showcasing DYG's capabilities are truly remarkable.
We're reaching out because we believe that hosting the pre-trained DYG model on Hugging Face (https://huggingface.co/models) would significantly benefit your research by boosting its visibility and accessibility within the broader AI community. Hugging Face provides tools for easy sharing, version control, and discoverability via detailed model cards and metadata tagging. We can also link your model directly to your Hugging Face Paper Page (if you choose to submit the paper to hf.co/papers).
We'd be happy to assist with the upload process, offering guidance and support as needed. If your model is built with PyTorch, you can take advantage of the PyTorchModelHubMixin for seamless integration with from_pretrained and push_to_hub functionality. We also have a ZeroGPU grant program (https://huggingface.co/docs/hub/en/spaces-gpus#community-gpu-grants) offering free A100 GPU access for building demos on Hugging Face Spaces.
Would you be open to considering this opportunity?
Kind regards,
Niels ML Engineer @ Hugging Face 🤗
Hi @NielsRogge
Thank you very much for your interest. Currently, our work is under review. Once it is accepted, we will immediately open-source the code. At that time, we will consider contacting you for assistance with the deployment on Hugging Face.
Thank you so much.
Great! I've already indexed your paper here: https://huggingface.co/papers/2501.18672.
Feel free to claim it as an author, and you can also add a Github URL