🌍 Guessing where an image was taken is a hard, and often ambiguous problem. Introducing diffusion-based geolocation—we predict global locations by refining random guesses into trajectories across the Earth's surface! . 🗺️ Paper, code, and demo:

RT @TimDarcet: @chrisoffner3d There's an irreducible error, so 99.99 is probably impossible.However, like in LLMs, we can keep scraping poi….

@AIatMeta @p_bojanowski Plonk project page: (@VickyKalogeiton,@david_picard,@loiclandrieu).

Congrats to @AIatMeta and @p_bojanowski team for the DinoV3 release! . Seeing it outperforms CLIP on "cultural knowledge" based task like geoloc make me very hopeful for it working really well in VLMs!.

🌍 Geoloc is a fantastic downstream benchmark:. - Requires fine-grained visual understanding (textures, vegetation, road signs, architecture). - Tests global generalization . - Forces models to pick up real-world cues. That’s why DinoV3 shining here is such a big deal 🚀.

Even crazier 🤯 DinoV3 works in some out-of-distribution setups too — as long as there are geographical cues 🌄🗺️. (Remember: the network is trained only on road images!). Where DinoV2 totally failed, DinoV3 is holding up 👊

The setup 👉 We use our riemannian flow matching model PLONK (CVPR25: Around the World in 80 Timesteps: A Generative Approach to Global Visual Geolocation) 🌍. We simply swap StreetCLIP with DinoV3 as a drop-in backbone, and train on OpenStreetView-5M. And boom 💥 — DinoV3 wins.

🚀 DinoV3 just became the new go-to backbone for geoloc!.It outperforms CLIP-like models (SigLip2, finetuned StreetCLIP)… and that’s shocking 🤯.Why? CLIP models have an innate advantage — they literally learn place names + images. DinoV3 doesn’t.

I had the privilege to be invited to speak about our work "Around the World in 80 Timesteps" at the French Podcast @UnderscoreTalk ! If you speak french, check it out!. If you want to learn more

RT @iScienceLuvr: Diffusion Beats Autoregressive in Data-Constrained Settings. Comparison of diffusion and autoregressive language models f….

RT @JunyuXieArthur: Movies are more than just video clips, they are stories! 🎬. We’re hosting the 1st SLoMO Workshop at #ICCV2025 to discus….

Come see us in poster 186 to see our poster Around the World in 80 timesteps: A generative Approach to Global Visual Geolocation!

RT @nico_dufour: 🌍 Guessing where an image was taken is a hard, and often ambiguous problem. Introducing diffusion-based geolocation—we pre….

RT @antoine_guedon: I'm at #CVPR2025 to present our paper 🍵MAtCha Gaussians🍵, today Friday afternoon, Hall D, Poster 53!. If you're in Nas….

RT @nico_dufour: I will be at #CVPR2025 this week in Nashville. I will be presenting our paper "Around the World in 80 Timesteps:.A Genera….

RT @g_astruc: 🛰️ At #CVPR2025 presenting "AnySat: An Earth Observation Model for Any Resolutions, Scales, and Modalities" - Saturday aftern….

We will present it at:. "3rd Workshop on Generative Models for Computer Vision" Wednesday 11 of June, 1pm. Main Conference: Sunday 15 of June, 10h30am, Poster 186.

I will be at #CVPR2025 this week in Nashville. I will be presenting our paper "Around the World in 80 Timesteps:.A Generative Approach to Global Visual Geolocation". We tackle geolocalization as a generative task allowing for SOTA performance and more interpretable predictions.

RT @nico_dufour: @pabloppp So in my experience, At this small scale, textual adherence is actually the "easiest" to have. We worked at thos….

RT @nico_dufour: @pabloppp You can check We train a 330M params model for around 500 H100 hours. I've been moderniz….

RT @pabloppp: What is a reasonable amount of GPU hours to train to convergence a "small" t2i diffusion model? 🤔 What would be considered gr….