search

heitorrapela / ModTr

[ECCV2024] ModTr: Modality Translation for Object Detection Adaptation Without Forgetting Prior Knowledge
https://arxiv.org/abs/2404.01492
MIT License
9 stars 0 forks source link

Open-sourcing pre-trained models #3

Open merveenoyan opened 1 month ago

merveenoyan commented 1 month ago

Hello @heitorrapela 🤗

I'm Merve and I work on computer vision at Hugging Face. I see your paper is accepted, congratulations on getting accepted to ECCV! I indexed it here https://huggingface.co/papers/2404.01492 at paper pages. The paper page lets people discuss about your paper and lets them find artifacts about it, you can also claim the paper as yours which will show up on your public profile at HF.

Would you like to host the model you've pre-trained with this technique on https://huggingface.co/models? I couldn't find the pre-trained models, are you planning to release them?

After uploaded, we can also link the datasets to the paper page (read here) so people can discover your model (and also we can grant you hardware if you decide to build a demo) What do you think?

heitorrapela commented 1 month ago

Hi @merveenoyan, thanks for the kind words and effort with the open-source community. 🤗

Sorry for the late response; I was traveling for the conference.

I am planning to release the weights, and I can try to release a demo as well.

I will have more time after the conference to do everything following some good practices. But I am open to doing it :)

Best Regards, Heitor R. Medeiros

merveenoyan commented 1 month ago

@heitorrapela sure! I will try to find you in person as well, currently in ECCV :)

heitorrapela commented 1 month ago

Hi @merveenoyan, yes sure we can meet in ECCV 👍🏻