BigEarthNet

Dataset Information
Modalities
Images, Hyperspectral images
License
Custom
Homepage
Official Website
Contents

Overview

BigEarthNet consists of 590,326 Sentinel-2 image patches, each of which is a section of i) 120x120 pixels for 10m bands; ii) 60x60 pixels for 20m bands; and iii) 20x20 pixels for 60m bands.

Source: BigEarthNet: A Large-Scale Benchmark Archive For Remote Sensing Image Understanding

Variants: BigEarthNet, BigEarthNet (official test set), BigEarthNet-S1 (official test set)

Associated Benchmarks

This dataset is used in 1 benchmark:

Recent Benchmark Submissions

Task Model Paper Date
Multi-Label Image Classification DINO-MC Extending global-local view alignment for … 2023-03-12
Multi-Label Image Classification MoCo-v2 (ResNet50, fine tune) SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal … 2022-11-13
Multi-Label Image Classification MoCo-v3 (ViT-S/16, fine tune) SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal … 2022-11-13
Multi-Label Image Classification MAE (ViT-S/16, fine tune) SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal … 2022-11-13
Multi-Label Image Classification MoCo-v2 (ResNet18, fine tune) Self-supervised Learning in Remote Sensing: … 2022-06-27
Multi-Label Image Classification MLPMixer Benchmarking and scaling of deep … 2021-11-18
Multi-Label Image Classification WideResNet-B5-ECA Benchmarking and scaling of deep … 2021-11-18
Multi-Label Image Classification ViTM/20 Benchmarking and scaling of deep … 2021-11-18
Multi-Label Image Classification ResNet50 Benchmarking and scaling of deep … 2021-11-18
Multi-Label Image Classification ResNet50 In-domain representation learning for remote … 2019-11-15

Research Papers

Recent papers with results on this dataset:

External Links: