Guillaume Astruc

🤔 What if embedding multimodal EO data was as easy as using a ResNet on images? Introducing AnySat: one model for any resolution (0.2m–250m), scale (0.3–2600 hectares), and modalities (choose from 11 sensors & time series)! Try it with just a few lines of code:

🛰️ At #CVPR2025 presenting "AnySat: An Earth Observation Model for Any Resolutions, Scales, and Modalities" - Saturday afternoon, Poster 355! If you're here and want to discuss geolocation or geospatial foundation models, let's connect!

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.

📢New preprint “When majority rules, minority loses: bias amplification of gradient descent” We often blame biased data but training also amplifies biases. Our paper explores how ML algorithms favor stereotypes at the expense of minority groups. ➡️arxiv.org/abs/2505.13122 (1/3)

We've added new experiments demonstrating robust generalization capabilities! Notably, AnySat shows strong performance on HLS Burn Scars - a sensor never seen during pretraining! 🔥🛰️ Check it out: 📄 Paper: arxiv.org/abs/2412.14123 🌐 Project: gastruc.github.io/anysat

Introducing Chapter-Llama [#CVPR2025], a framework for 𝐯𝐢𝐝𝐞𝐨 𝐜𝐡𝐚𝐩𝐭𝐞𝐫𝐢𝐧𝐠 using Large Language Models! 🎬🦙 Check it out: 📄 Paper: arxiv.org/abs/2504.00072 🔗 Project: imagine.enpc.fr/~lucas.ventura… 💻 Code: github.com/lucas-ventura/… 🤗 Demo: huggingface.co/spaces/lucas-v…

💻We've released the code for our #CVPR2025 paper MAtCha! 🍵MAtCha reconstructs sharp, accurate and scalable meshes of both foreground AND background from just a few unposed images (eg 3 to 10 images)... ...While also working with dense-view datasets (hundreds of images)!

1/13 🐊 Introducing our latest work on improving relative camera pose regression with a novel pre-training approach Alligat0R (arxiv.org/abs/2503.07561)! @GBourmaud @VincentLepetit2

🧩Excited to share our paper "RUBIK: A Structured Benchmark for Image Matching across Geometric Challenges" arxiv.org/abs/2502.19955 accepted to #CVPR2025! We created a benchmark that systematically evaluates image matching methods across well-defined geometric difficulty levels.

🎉 Excited to share my first single-author preprint. Proud of this contribution. I would be grateful for any feedback or thoughts. 🙌

AnySat: An Earth Observation Model for Any Resolutions, Scales, and Modalities

AnySat: An Earth Observation Model for Any Resolutions, Scales, and Modalities @g_astruc @NicaoGr Clement Mallet @loiclandrieu tldr: pretrains a multisensor & multimodal joint embedding network on GeoPlex dataset. achieves SOTA on a diverse set of time-series EO benchmarks

⚠️Reconstructing sharp 3D meshes from a few unposed images is a hard and ambiguous problem. ☑️With MAtCha, we leverage a pretrained depth model to recover sharp meshes from sparse views including both foreground and background, within mins!🧵 🌐Webpage: anttwo.github.io/matcha/

🌍 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: nicolas-dufour.github.io/plonk

🚀 Excited to share our new preprint with @jerome_bolte, E. Pauwels & A. Purica: "A second-order-like optimizer with adaptive gradient scaling for deep learning." Check out the paper: arxiv.org/pdf/2410.05871 GitHub: github.com/innaprop/innap… (1/3)