# RSPDBench: Benchmarking Vision Foundation Models on Earth Observation Tasks Under Physically Grounded Remote-Sensing Product Degradations

arXiv: https://arxiv.org/abs/2609.23427 · HTML: https://mlsanity.com/paper/2609.23427/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 37 references that are in the map.
- First posted on arXiv 20 Sep 2026.

## Builds on (most cited first, 19 of 19)

- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [Scale-MAE: A Scale-Aware Masked Autoencoder for Multiscale Geospatial Representation Learning](https://mlsanity.com/paper/2212.14532/) (2022)
- [SkySense: A Multi-Modal Remote Sensing Foundation Model Towards Universal Interpretation for Earth Observation Imagery](https://mlsanity.com/paper/2312.10115/) (2023)
- [CROMA: Remote Sensing Representations with Contrastive Radar-Optical Masked Autoencoders](https://mlsanity.com/paper/2311.00566/) (2023)
- [Prithvi-EO-2.0: A Versatile Multi-Temporal Foundation Model for Earth Observation Applications](https://mlsanity.com/paper/2412.02732/) (2024)
- [GEO-Bench: Toward Foundation Models for Earth Monitoring](https://mlsanity.com/paper/2306.03831/) (2023)
- [TerraMind: Large-Scale Generative Multimodality for Earth Observation](https://mlsanity.com/paper/2504.11171/) (2025)
- [Rethinking Transformers Pre-training for Multi-Spectral Satellite Imagery](https://mlsanity.com/paper/2403.05419/) (2024)
- [Towards a Unified Copernicus Foundation Model for Earth Vision](https://mlsanity.com/paper/2503.11849/) (2025)
- [Galileo: Learning Global & Local Features of Many Remote Sensing Modalities](https://mlsanity.com/paper/2502.09356/) (2025)
- [Panopticon: Advancing Any-Sensor Foundation Models for Earth Observation](https://mlsanity.com/paper/2503.10845/) (2025)
- [CrossEarth: Geospatial Vision Foundation Model for Domain Generalizable Remote Sensing Semantic Segmentation](https://mlsanity.com/paper/2410.22629/) (2024)
- [PhilEO Bench: Evaluating Geo-Spatial Foundation Models](https://mlsanity.com/paper/2401.04464/) (2024)
- [GEO-Bench-2: From Performance to Capability, Rethinking Evaluation in Geospatial AI](https://mlsanity.com/paper/2511.15658/) (2025)
- [REOBench: Benchmarking Robustness of Earth Observation Foundation Models](https://mlsanity.com/paper/2505.16793/) (2025)
- [Unlocking the Use of Raw Multispectral Earth Observation Imagery for Onboard Artificial Intelligence](https://mlsanity.com/paper/2305.11891/) (2023)
- [EarthShift: a benchmark for measuring robustness to real-world distribution shifts in Earth observation](https://mlsanity.com/paper/2605.29330/) (2026)
- [GeoCrossBench: Cross-Band Generalization for Remote Sensing](https://mlsanity.com/paper/2511.02831/) (2025)
- [A Unified Foundation Model for All-in-One Multi-Modal Remote Sensing Image Restoration and Fusion with Language Prompting](https://mlsanity.com/paper/2604.05629/) (2026)

## Abstract (arXiv)

Vision foundation models targeting Earth observation (EO) tasks are commonly evaluated on clean downstream benchmarks, but operational EO products can already contain spatial, radiometric, alignment, noise, and harmonization defects before reaching the model. Existing robustness evaluations often use generic image corruptions or broad domain shifts, which do not isolate these product-level failure modes. We introduce RSPDBench, a physically grounded remote-sensing-product degradation benchmark for vision foundation models. RSPDBench evaluates five EO datasets, seven foundation-model entries, and two supervised baselines under audited primitive degradations and compound product chains. Each model is evaluated under its clean-selected native protocol, with robustness measured as the drop from its own clean baseline. Our analysis reveals that degradation sensitivity is strongly structured: resolution-conditioned and channel-grouped encoders protect different failure axes, and the same physical defect can hurt one model while helping another. Compound chains expose failures that isolated degradations do not predict, with model-dependent amplification, saturation, or component dominance, and excess drops up to 38 percentage points beyond the strongest component. These results show that EO robustness cannot be characterized by clean accuracy or generic perturbation tests alone; it must also be measured against the structured defects that remote-sensing products carry into deployment.