FLORO: A Multimodal Geospatial Foundation Model for Ecological Remote Sensing Across Sensors and Scales
cs.CV, cs.AI
Submitted: 2026-05-27
Updated: 2026-10-29
Comments: 57 pages, 11 figures
Journal ref: ISPRS Journal of Photogrammetry and Remote Sensing 243, Part 1 (2027) 811-839
DOI: 10.1016/j.isprsjprs.2026.10.008
Project page: https://jorlrodriguezg.github.io/floro
License: http://creativecommons.org/licenses/by/4.0/
The gist: Foundation models offer a promising route to transferable remote sensing representations, but many current approaches depend on very large pretraining datasets and fixed sensor configurations…
Terminology
Abstract
Foundation models offer a promising route to transferable remote sensing representations, but many current approaches depend on very large pretraining datasets and fixed sensor configurations, limiting their suitability for ecological and environmental applications, where observations often vary across platforms, spatial and spectral resolutions, and available modalities. We introduce FLORO, a multimodal geospatial foundation model designed to learn transferable representations from a small but highly diverse remote sensing corpus. FLORO is pretrained using masked autoencoding on a heterogeneous combination of Sentinel-1, Sentinel-2, SkySAT imagery, elevation, and UAV-derived data. To accommodate sensor variability, FLORO incorporates availability-aware inputs that indicate which spectral bands and auxiliary modalities are present in each sample, enabling a unified input space across heterogeneous sensor configurations. We evaluated FLORO on the PANGAEA benchmark under a frozen-encoder protocol across scene classification, segmentation, and regression tasks. Despite being pretrained on a smaller corpus than competing foundation models, FLORO achieved strong and stable transfer across optical, optical-SAR, and optical-elevation benchmarks spanning medium-resolution satellite, airborne, and ultra-high-resolution UAV imagery. FLORO obtained the second-best average segmentation performance across six PANGAEA benchmarks, the third-best average rank across three scene-classification benchmarks, and excelled in regression tasks, while qualitative results showed improved preservation of spatial structure in flood, urban, biomass, and canopy-height prediction settings. Overall, these results position FLORO as a practical framework for ecological and environmental monitoring under heterogeneous sensing conditions.
Sources
- MultiMAE: Multi-modal Multi-task Masked Autoencoders
- SatlasPretrain: A Large-Scale Dataset for Remote Sensing Image Understanding
- On the Opportunities and Risks of Foundation Models
- LightFormer: A lightweight and efficient decoder for remote sensing image segmentation
- SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- CROMA: Remote Sensing Representations with Contrastive Radar-Optical Masked Autoencoders
- Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
- EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification
- Foundation Models in Remote Sensing: Evolving from Unimodality to Multimodality
- Masked Autoencoders that Listen
- Foundation Models for Generalist Geospatial Artificial Intelligence
- TerraMind: Large-Scale Generative Multimodality for Earth Observation
- RemoteCLIP: A Vision Language Foundation Model for Remote Sensing
- Decoupled Weight Decay Regularization
- Vision Foundation Models in Remote Sensing: A Survey
- PANGAEA: A Global and Inclusive Benchmark for Geospatial Foundation Models
- Fine-tune Smarter, Not Harder: Parameter-Efficient Fine-Tuning for Geospatial Foundation Models
- Towards Geospatial Foundation Models via Continual Pretraining
- Mixed Precision Training
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs