Landsat-Bench: Datasets and Benchmarks for Landsat Foundation Models
By: Isaac Corley, Lakshay Sharma, Ruth Crasto
Potential Business Impact:
Helps computers understand Earth pictures better.
The Landsat program offers over 50 years of globally consistent Earth imagery. However, the lack of benchmarks for this data constrains progress towards Landsat-based Geospatial Foundation Models (GFM). In this paper, we introduce Landsat-Bench, a suite of three benchmarks with Landsat imagery that adapt from existing remote sensing datasets -- EuroSAT-L, BigEarthNet-L, and LC100-L. We establish baseline and standardized evaluation methods across both common architectures and Landsat foundation models pretrained on the SSL4EO-L dataset. Notably, we provide evidence that SSL4EO-L pretrained GFMs extract better representations for downstream tasks in comparison to ImageNet, including performance gains of +4% OA and +5.1% mAP on EuroSAT-L and BigEarthNet-L.
Similar Papers
Mars-Bench: A Benchmark for Evaluating Foundation Models for Mars Science Tasks
CV and Pattern Recognition
Tests AI for understanding Mars pictures.
GEO-Bench-2: From Performance to Capability, Rethinking Evaluation in Geospatial AI
CV and Pattern Recognition
Helps pick the best AI for looking at Earth.
Towards a Unified Copernicus Foundation Model for Earth Vision
CV and Pattern Recognition
Lets satellites understand Earth better, from land to air.