# Fine-tuning of Geospatial Foundation Models for Aboveground Biomass Estimation

arXiv: https://arxiv.org/abs/2406.19888 · HTML: https://mlsanity.com/paper/2406.19888/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 33 references that are in the map.
- First posted on arXiv 28 Jun 2024.

## Builds on (most cited first, 17 of 17)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Masked-attention Mask Transformer for Universal Image Segmentation](https://mlsanity.com/paper/2112.01527/) (2021)
- [Unified Perceptual Parsing for Scene Understanding](https://mlsanity.com/paper/1807.10221/) (2018)
- [Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network](https://mlsanity.com/paper/1609.05158/) (2016)
- [SimMIM: A Simple Framework for Masked Image Modeling](https://mlsanity.com/paper/2111.09886/) (2021)
- [DOTA: A Large-scale Dataset for Object Detection in Aerial Images](https://mlsanity.com/paper/1711.10398/) (2017)
- [DeepGlobe 2018: A Challenge to Parse the Earth through Satellite Images](https://mlsanity.com/paper/1805.06561/) (2018)
- [Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Data](https://mlsanity.com/paper/2103.16607/) (2021)
- [iSAID: A Large-scale Dataset for Instance Segmentation in Aerial Images](https://mlsanity.com/paper/1905.12886/) (2019)
- [Foundation Models for Generalist Geospatial Artificial Intelligence](https://mlsanity.com/paper/2310.18660/) (2023)
- [Global canopy height regression and uncertainty estimation from GEDI LIDAR waveforms with deep ensembles](https://mlsanity.com/paper/2103.03975/) (2021)
- [A General Purpose Neural Architecture for Geospatial Systems](https://mlsanity.com/paper/2211.02348/) (2022)
- [Aboveground carbon biomass estimate with Physics-informed deep network](https://mlsanity.com/paper/2210.13752/) (2022)
- [A generic self-supervised learning (SSL) framework for representation learning from spectra-spatial feature of unlabeled remote sensing imagery](https://mlsanity.com/paper/2306.15836/) (2023)

## Built on by (most cited first, 3 of 3)

- [LiDAR Remote Sensing Meets Weak Supervision: Concepts, Methods, and Perspectives](https://mlsanity.com/paper/2503.18384/) (2025)
- [HyBiomass: Global Hyperspectral Imagery Benchmark Dataset for Evaluating Geospatial Foundation Models in Forest Aboveground Biomass Estimation](https://mlsanity.com/paper/2506.11314/) (2025)
- [Revisiting Learning Rate Control](https://mlsanity.com/paper/2507.01724/) (2025)

## Abstract (arXiv)

Global vegetation structure mapping is critical for understanding the global carbon cycle and maximizing the efficacy of nature-based carbon sequestration initiatives. Moreover, vegetation structure mapping can help reduce the impacts of climate change by, for example, guiding actions to improve water security, increase biodiversity and reduce flood risk. Global satellite measurements provide an important set of observations for monitoring and managing deforestation and degradation of existing forests, natural forest regeneration, reforestation, biodiversity restoration, and the implementation of sustainable agricultural practices. In this paper, we explore the effectiveness of fine-tuning of a geospatial foundation model to estimate above-ground biomass (AGB) using space-borne data collected across different eco-regions in Brazil. The fine-tuned model architecture consisted of a Swin-B transformer as the encoder (i.e., backbone) and a single convolutional layer for the decoder head. All results were compared to a U-Net which was trained as the baseline model Experimental results of this sparse-label prediction task demonstrate that the fine-tuned geospatial foundation model with a frozen encoder has comparable performance to a U-Net trained from scratch. This is despite the fine-tuned model having 13 times less parameters requiring optimization, which saves both time and compute resources. Further, we explore the transfer-learning capabilities of the geospatial foundation models by fine-tuning on satellite imagery with sparse labels from different eco-regions in Brazil.