OceanMAE: A Foundation Model for Ocean Remote Sensing
2026-04-09 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors created OceanMAE, a tool that helps computers better understand ocean images by using special ocean-related information during learning. Unlike typical models trained mostly on land images, theirs learns from many unlabeled ocean pictures and extra ocean data to improve how well it detects things like marine debris and estimates underwater shapes. They tested OceanMAE on different tasks and found it works especially well for identifying marine pollution. Their work shows that adding ocean-specific knowledge during training can make ocean remote sensing models perform better.
ocean remote sensingmasked autoencoderself-supervised learningSentinel-2marine segmentationbathymetry estimationmultispectral imagerymarine debris detectionHydro datasetUNet
Authors
Viola-Joanna Stamer, Panagiotis Agrafiotis, Behnood Rasti, Begüm Demir
Abstract
Accurate ocean mapping is essential for applications such as bathymetry estimation, seabed characterization, marine litter detection, and ecosystem monitoring. However, ocean remote sensing (RS) remains constrained by limited labeled data and by the reduced transferability of models pre-trained mainly on land-dominated Earth observation imagery. In this paper, we propose OceanMAE, an ocean-specific masked autoencoder that extends standard MAE pre-training by integrating multispectral Sentinel-2 observations with physically meaningful ocean descriptors during self-supervised learning. By incorporating these auxiliary ocean features, OceanMAE is designed to learn more informative and ocean-aware latent representations from large- scale unlabeled data. To transfer these representations to downstream applications, we further employ a modified UNet-based framework for marine segmentation and bathymetry estimation. Pre-trained on the Hydro dataset, OceanMAE is evaluated on MADOS and MARIDA for marine pollutant and debris segmentation, and on MagicBathyNet for bathymetry regression. The experiments show that OceanMAE yields the strongest gains on marine segmentation, while bathymetry benefits are competitive and task-dependent. In addition, an ablation against a standard MAE on MARIDA indicates that incorporating auxiliary ocean descriptors during pre-training improves downstream segmentation quality. These findings highlight the value of physically informed and domain-aligned self-supervised pre- training for ocean RS. Code and weights are publicly available at https://git.tu-berlin.de/joanna.stamer/SSLORS2.