← Back to all news
arXiv Computer Vision By Miro Miranda, Rocco Sedona, Valerio Marsocci, Gabriele Cavallaro, Claudia Paris

Ground-to-Satellite Knowledge Distillation for Crop Classification under Temporal and Geographic Shift

Read the original on arXiv Computer Vision →

Ground-to-satellite knowledge distillation leverages ground-level field photographs to provide privileged supervision for crop classification models built on pretrained satellite embeddings. Using a frozen CLIP-based photo teacher, classifiers trained on AlphaEarth and TESSERA representations achieve consistent improvements over satellite-only baselines, especially under temporal and geographic shifts. The method enhances robustness without altering inference requirements, as photographs are used only during training.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

Related stories

arXiv Computer Vision
Sep 24

AgroBench: A Reproducible Multimodal Benchmark for Weakly Supervised Crop Yield Learning from County Statistics and Pixel Observations

AgroBench is a reproducible benchmark that converts U.S. county-level crop yield statistics into weakly supervised pixel‑level crop time series. The data generation pipeline fuses USDA yield data with land cover masks, Sentinel‑2 and Sentinel‑1 imagery, climatic variables, and terrain information to produce multimodal sequences for individual crop pixels across the growing season. The benchmark includes over 13 million observations from 788,654 crop pixels, covering 5,107 county‑year combinations for five major U.S. crops from 2017 to 2024, and establishes a Leave‑One‑Year‑Out evaluation protocol with baseline machine learning results.

By Udaiveer Singh, Rajiv Ranjan, Shashank Tamaskar, Dharmendra Saraswat
multimodalbenchmarks
More like this →
arXiv AI
Sep 1

A Composition-Aware Pretraining Framework for Geospatial Foundation Models

The paper introduces a composition‑aware pretraining framework for geospatial foundation models that explicitly encodes fractional land‑cover mixtures as histogram targets for each satellite image cell. By using Earth Mover’s Distance to distill these composition targets into a 36.8 M‑parameter backbone, the authors demonstrate significant improvements on region‑level tasks such as zero‑shot image retrieval and scene classification, while maintaining competitive performance on fine‑grained tasks like segmentation and object detection. The method outperforms larger models (SatMAE and Prithvi‑EO‑2.0) and achieves a 55.6 % relative boost on the ForestNet‑12 dataset, evidencing the benefit of explicit composition modeling.

By Aryan Kashyap Naveen, Abhishek Srinivas, Pranav Moothedath, Shrutilipi Bhattacharjee
computer-visionbenchmarks
More like this →
arXiv Computer Vision
Sep 24

VLM2GeoVec: Toward Universal Multimodal Embeddings for Remote Sensing

The paper introduces RSMEB, a unified benchmark for remote‑sensing multimodal retrieval that evaluates both cross‑modal and interleaved retrieval across 21 tasks under a single ranking protocol. It also presents VLM2GeoVec, an instruction‑conditioned single‑encoder model that embeds image, text, bounding‑box, and geo‑coordinate tokens into one sequence and achieves state‑of‑the‑art performance on region‑caption, referring‑expression, and semantic geo‑aware retrieval while remaining competitive on conventional tasks. The authors provide code, checkpoints, and data on GitHub to facilitate reproducibility.

By Emanuel S\'anchez Aimar, Gulnaz Zhambulova, Fahad Shahbaz Khan, Yonghao Xu, Michael Felsberg
ragmultimodalbenchmarks
More like this →