UniverSat: Resolution- and Modality-Agnostic Transformers for Earth Observation

Hugging Face Daily Papers Papers

Summary

UniverSat introduces a Universal Patch Encoder for Vision Transformers that enables robust, sensor-agnostic spatial feature extraction across diverse Earth Observation data types, achieving strong results on classification and segmentation benchmarks.

Vision Transformers (ViT) dominate computer vision. However, their reliance on rigid patch projectors hinders transfer to Earth Observation (EO), where input modalities, scales, and resolutions vary widely. We introduce UniverSat, a ViT-style backbone built around a Universal Patch Encoder that maps patches from arbitrary spatial, spectral, and temporal resolutions, and from both optical and non-optical sensors, into a shared embedding space with a shared set of weights. This enables training a single model on heterogeneous multimodal corpora via self-supervision, yielding robust, sensor-agnostic spatial features. We validate this approach with strong results across classification and segmentation on standard EO benchmarks from GeoBench, PANGEABench, and SpectralEarth. Our code and models are available at https://github.com/gastruc/UniverSat.
Original Article
View Cached Full Text

Cached at: 06/23/26, 05:40 AM

Paper page - UniverSat: Resolution- and Modality-Agnostic Transformers for Earth Observation

Source: https://huggingface.co/papers/2606.23503

Abstract

UniverSat introduces a Universal Patch Encoder for Vision Transformers that enables robust, sensor-agnostic spatial feature extraction across diverse Earth Observation data types.

Vision Transformers(ViT) dominate computer vision. However, their reliance on rigidpatch projectorshinders transfer toEarth Observation(EO), where input modalities, scales, and resolutions vary widely. We introduce UniverSat, a ViT-style backbone built around aUniversal Patch Encoderthat maps patches from arbitrary spatial, spectral, and temporal resolutions, and from both optical and non-optical sensors, into a shared embedding space with a shared set of weights. This enables training a single model on heterogeneousmultimodal corporaviaself-supervision, yielding robust, sensor-agnosticspatial features. We validate this approach with strong results acrossclassificationandsegmentationon standard EO benchmarks from GeoBench, PANGEABench, and SpectralEarth. Our code and models are available at https://github.com/gastruc/UniverSat.

View arXiv pageView PDFProject pageGitHub1Add to collection

Models citing this paper1

#### g-astruc/UniverSat Image Feature Extraction• 0.2B• Updatedabout 1 hour ago • 5 • 3

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2606.23503 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.23503 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

ViT-Up: Faithful Feature Upsampling for Vision Transformers

Hugging Face Daily Papers

ViT-Up introduces a task-agnostic feature upsampler for Vision Transformers that predicts features at arbitrary continuous image coordinates, enabling dense feature maps at any resolution and improving dense prediction and semantic correspondence benchmarks. It outperforms prior state-of-the-art upsamplers, with gains of up to +2.07 mIoU on Cityscapes and +4.17 [email protected] on SPair-71k.

UniT: Unified Geometry Learning with Group Autoregressive Transformer

Hugging Face Daily Papers

UniT is a unified feed-forward model for geometry perception using a Group Autoregressive Transformer that integrates multiple paradigms (online/offline, multi-modal, long-horizon) while maintaining metric-scale accuracy via scale-adaptive loss and queue-style KV caching. It achieves state-of-the-art performance on ten benchmarks spanning seven tasks.