Ultra-high Spatial Resolution Land Cover Classification is essential for fine-grained land cover analysis, yet it remains challenging due to the high cost of pixel-level annotations, significant scale variation, and the limited adaptability of large-scale vision models. Existing methods typically focus on 1-meter spatial resolution imagery and rely heavily on annotated data, whereas practical applications often require processing higher-resolution imagery under weak supervision. To address this, we propose a parameter-efficient semi-supervised segmentation framework for 0.3 m spatial resolution imagery, which leverages the knowledge of SAM2 and introduces a remote sensing-specific FreqWeaver Adapter to enhance fine-grained detail modeling while maintaining a lightweight design at only 5.96% of the total model parameters. By effectively leveraging unlabeled data and maintaining minimal parameter overhead, the proposed method delivers robust segmentation results with superior structural consistency, achieving a 1.78% improvement over existing parameter-efficient tuning strategies and a 3.44% gain compared to state-of-the-art high-resolution remote sensing segmentation approaches.
Super-resolution aims to increase the resolution of satellite images by reconstructing high-frequency details, which go beyond na\"ive upsampling. This has particular relevance for Earth observation missions like Sentinel-2, which offer frequent, regular coverage at no cost; but at coarse resolution. Its pixel footprint is too large to capture small features like houses, streets, or hedge rows. To address this, we present SEN4X, a hybrid super-resolution architecture that combines the advantages of single-image and multi-image techniques. It combines temporal oversampling from repeated Sentinel-2 acquisitions with a learned prior from high-resolution Pl\'eiades Neo data. In doing so, SEN4X upgrades Sentinel-2 imagery to 2.5 m ground sampling distance. We test the super-resolved images on urban land-cover classification in Hanoi, Vietnam. We find that they lead to a significant performance improvement over state-of-the-art super-resolution baselines.
Multimodal remote sensing data, including spectral and lidar or photogrammetry, is crucial for achieving satisfactory land-use / land-cover classification results in urban scenes. So far, most studies have been conducted in a 2D context. When 3D information is available in the dataset, it is typically integrated with the 2D data by rasterizing the 3D data into 2D formats. Although this method yields satisfactory classification results, it falls short in fully exploiting the potential of 3D data by restricting the model's ability to learn 3D spatial features directly from raw point clouds. Additionally, it limits the generation of 3D predictions, as the dimensionality of the input data has been reduced. In this study, we propose a fully 3D-based method that fuses all modalities within the 3D point cloud and employs a dedicated dual-branch Transformer model to simultaneously learn geometric and spectral features. To enhance the fusion process, we introduce a cross-attention-based mechanism that fully operates on 3D points, effectively integrating features from various modalities across multiple scales. The purpose of cross-attention is to allow one modality to assess the importance of another by weighing the relevant features. We evaluated our method by comparing it against both 3D and 2D methods using the 2018 IEEE GRSS Data Fusion Contest (DFC2018) dataset. Our findings indicate that 3D fusion delivers competitive results compared to 2D methods and offers more flexibility by providing 3D predictions. These predictions can be projected onto 2D maps, a capability that is not feasible in reverse. Additionally, we evaluated our method on different datasets, specifically the ISPRS Vaihingen 3D and the IEEE 2019 Data Fusion Contest. Our code will be published here: https://github.com/aldinorizaldy/hyperpointformer.
Accurate wetland land-cover classification is essential for environmental monitoring, biodiversity assessment, and sustainable ecosystem management. However, the scarcity of annotated data, especially for high-resolution satellite imagery, poses a significant challenge for supervised learning approaches. To tackle this issue, this study presents a methodology for wetland land-cover segmentation and classification that adopts both supervised and self-supervised learning (SSL). We train a U-Net model from scratch on Sentinel-2 imagery across six wetland regions in the Netherlands, achieving a baseline model accuracy of 85.26%. Addressing the limited availability of labeled data, the results show that SSL pretraining with an autoencoder can improve accuracy, especially for the high-resolution imagery where it is more difficult to obtain labeled data, reaching an accuracy of 88.23%. Furthermore, we introduce a framework to scale manually annotated high-resolution labels to medium-resolution inputs. While the quantitative performance between resolutions is comparable, high-resolution imagery provides significantly sharper segmentation boundaries and finer spatial detail. As part of this work, we also contribute a curated Sentinel-2 dataset with Dynamic World labels, tailored for wetland classification tasks and made publicly available.
We introduce Land-MoE, a novel approach for multispectral land cover classification (MLCC). Spectral shift, which emerges from disparities in sensors and geospatial conditions, poses a significant challenge in this domain. Existing methods predominantly rely on domain adaptation and generalization strategies, often utilizing small-scale models that exhibit limited performance. In contrast, Land-MoE addresses these issues by hierarchically inserting a Frequency-aware Mixture of Low-rank Token Experts, to fine-tune Vision Foundation Models (VFMs) in a parameter-efficient manner. Specifically, Land-MoE comprises two key modules: the mixture of low-rank token experts (MoLTE) and frequency-aware filters (FAF). MoLTE leverages rank-differentiated tokens to generate diverse feature adjustments for individual instances within multispectral images. By dynamically combining learnable low-rank token experts of varying ranks, it enhances the robustness against spectral shifts. Meanwhile, FAF conducts frequency-domain modulation on the refined features. This process enables the model to effectively capture frequency band information that is strongly correlated with semantic essence, while simultaneously suppressing frequency noise irrelevant to the task. Comprehensive experiments on MLCC tasks involving cross-sensor and cross-geospatial setups demonstrate that Land-MoE outperforms existing methods by a large margin. Additionally, the proposed approach has also achieved state-of-the-art performance in domain generalization semantic segmentation tasks of RGB remote sensing images.
Hyperspectral imagery provides rich spectral detail but poses unique challenges because of its high dimensionality in both spatial and spectral domains. We propose \textit{HyperspectralMAE}, a Transformer-based foundation model for hyperspectral data that employs a \textit{dual masking} strategy: during pre-training we randomly occlude 50\% of spatial patches and 50\% of spectral bands. This forces the model to learn representations capable of reconstructing missing information across both dimensions. To encode spectral order, we introduce learnable harmonic Fourier positional embeddings based on wavelength. The reconstruction objective combines mean-squared error (MSE) with the spectral angle mapper (SAM) to balance pixel-level accuracy and spectral-shape fidelity. The resulting model contains about $1.8\times10^{8}$ parameters and produces 768-dimensional embeddings, giving it sufficient capacity for transfer learning. We pre-trained HyperspectralMAE on two large hyperspectral corpora -- NASA EO-1 Hyperion ($\sim$1\,600 scenes, $\sim$$3\times10^{11}$ pixel spectra) and DLR EnMAP Level-0 ($\sim$1\,300 scenes, $\sim$$3\times10^{11}$ pixel spectra) -- and fine-tuned it for land-cover classification on the Indian Pines benchmark. HyperspectralMAE achieves state-of-the-art transfer-learning accuracy on Indian Pines, confirming that masked dual-dimensional pre-training yields robust spectral-spatial representations. These results demonstrate that dual masking and wavelength-aware embeddings advance hyperspectral image reconstruction and downstream analysis.
Multi-source remote sensing data joint classification aims to provide accuracy and reliability of land cover classification by leveraging the complementary information from multiple data sources. Existing methods confront two challenges: inter-frequency multi-source feature coupling and inconsistency of complementary information exploration. To solve these issues, we present a Prototype-based Information Compensation Network (PICNet) for land cover classification based on HSI and SAR/LiDAR data. Specifically, we first design a frequency interaction module to enhance the inter-frequency coupling in multi-source feature extraction. The multi-source features are first decoupled into high- and low-frequency components. Then, these features are recoupled to achieve efficient inter-frequency communication. Afterward, we design a prototype-based information compensation module to model the global multi-source complementary information. Two sets of learnable modality prototypes are introduced to represent the global modality information of multi-source data. Subsequently, cross-modal feature integration and alignment are achieved through cross-attention computation between the modality-specific prototype vectors and the raw feature representations. Extensive experiments on three public datasets demonstrate the significant superiority of our PICNet over state-of-the-art methods. The codes are available at https://github.com/oucailab/PICNet.
The fusion of Synthetic Aperture Radar (SAR) and RGB imagery for land cover classification remains challenging due to modality heterogeneity and the underutilization of spectral complementarity. Existing methods often fail to decouple shared structural features from modality-specific radiometric attributes, leading to feature conflicts and information loss. To address this issue, we propose Phase-Amplitude Decoupling (PAD), a frequency-aware framework that separates phase (modality-shared) and amplitude (modality-specific) components in the Fourier domain. Specifically, PAD consists of two key components: 1) Phase Spectrum Correction (PSC), which aligns cross-modal phase features through convolution-guided scaling to enhance geometric consistency, and 2) Amplitude Spectrum Fusion (ASF), which dynamically integrates high-frequency details and low-frequency structures using frequency-adaptive multilayer perceptrons. This approach leverages SAR's sensitivity to morphological features and RGB's spectral richness. Extensive experiments on WHU-OPT-SAR and DDHR-SK datasets demonstrate state-of-the-art performance. Our work establishes a new paradigm for physics-aware multi-modal fusion in remote sensing. The code will be available at https://github.com/RanFeng2/PAD.
The increasing accessibility of remotely sensed data and the potential of such data to inform large-scale decision-making has driven the development of deep learning models for many Earth Observation tasks. Traditionally, such models must be trained on large datasets. However, the common assumption that broadly larger datasets lead to better outcomes tends to overlook the complexities of the data distribution, the potential for introducing biases and noise, and the computational resources required for processing and storing vast datasets. Therefore, effective solutions should consider both the quantity and quality of data. In this paper, we propose six novel core-set selection methods for selecting important subsets of samples from remote sensing image segmentation datasets that rely on imagery only, labels only, and a combination of each. We benchmark these approaches against a random-selection baseline on three commonly used land cover classification datasets: DFC2022, Vaihingen, and Potsdam. In each of the datasets, we demonstrate that training on a subset of samples outperforms the random baseline, and some approaches outperform training on all available data. This result shows the importance and potential of data-centric learning for the remote sensing domain. The code is available at https://github.com/keillernogueira/data-centric-rs-classification/.
Land use and land cover mapping from Earth Observation (EO) data is a critical tool for sustainable land and resource management. While advanced machine learning and deep learning algorithms excel at analyzing EO imagery data, they often overlook crucial geospatial metadata information that could enhance scalability and accuracy across regional, continental, and global scales. To address this limitation, we propose BRIDGE-LC (Bi-level Representation Integration for Disentangled GEospatial Land Cover), a novel deep learning framework that integrates multi-scale geospatial information into the land cover classification process. By simultaneously leveraging fine-grained (latitude/longitude) and coarse-grained (biogeographical region) spatial information, our lightweight multi-layer perceptron architecture learns from both during training but only requires fine-grained information for inference, allowing it to disentangle region-specific from region-agnostic land cover features while maintaining computational efficiency. To assess the quality of our framework, we use an open-access in-situ dataset and adopt several competing classification approaches commonly considered for large-scale land cover mapping. We evaluated all approaches through two scenarios: an extrapolation scenario in which training data encompasses samples from all biogeographical regions, and a leave-one-region-out scenario where one region is excluded from training. We also explore the spatial representation learned by our model, highlighting a connection between its internal manifold and the geographical information used during training. Our results demonstrate that integrating geospatial information improves land cover mapping performance, with the most substantial gains achieved by jointly leveraging both fine- and coarse-grained spatial information.
Earth observation satellites like Sentinel-1 (S1) and Sentinel-2 (S2) provide complementary remote sensing (RS) data, but S2 images are often unavailable due to cloud cover or data gaps. To address this, we propose a diffusion model (DM)-based approach for SAR-to-RGB translation, generating synthetic optical images from SAR inputs. We explore three different setups: two using Standard Diffusion, which reconstruct S2 images by adding and removing noise (one without and one with class conditioning), and one using Cold Diffusion, which blends S2 with S1 before removing the SAR signal. We evaluate the generated images in downstream tasks, including land cover classification and cloud removal. While generated images may not perfectly replicate real S2 data, they still provide valuable information. Our results show that class conditioning improves classification accuracy, while cloud removal performance remains competitive despite our approach not being optimized for it. Interestingly, despite exhibiting lower perceptual quality, the Cold Diffusion setup performs well in land cover classification, suggesting that traditional quantitative evaluation metrics may not fully reflect the practical utility of generated images. Our findings highlight the potential of DMs for SAR-to-RGB translation in RS applications where RGB images are missing.
The rapid expansion of multi-source satellite imagery drives innovation in Earth observation, opening unprecedented opportunities for Remote Sensing Foundation Models to harness diverse data. However, many existing models remain constrained by fixed spatial resolutions and patch sizes, limiting their ability to fully exploit the heterogeneous spatial characteristics inherent in satellite imagery. To address these challenges, we propose FlexiMo, a flexible remote sensing foundation model that endows the pre-trained model with the flexibility to adapt to arbitrary spatial resolutions. Central to FlexiMo is a spatial resolution-aware module that employs a parameter-free alignment embedding mechanism to dynamically recalibrate patch embeddings based on the input image's resolution and dimensions. This design not only preserves critical token characteristics and ensures multi-scale feature fidelity but also enables efficient feature extraction without requiring modifications to the underlying network architecture. In addition, FlexiMo incorporates a lightweight channel adaptation module that leverages prior spectral information from sensors. This mechanism allows the model to process images with varying numbers of channels while maintaining the data's intrinsic physical properties. Extensive experiments on diverse multimodal, multi-resolution, and multi-scale datasets demonstrate that FlexiMo significantly enhances model generalization and robustness. In particular, our method achieves outstanding performance across a range of downstream tasks, including scene classification, land cover classification, urban building segmentation, and cloud detection. By enabling parameter-efficient and physically consistent adaptation, FlexiMo paves the way for more adaptable and effective foundation models in real-world remote sensing applications.
The rapid advancement of remote sensing foundation models, particularly vision and multimodal models, has significantly enhanced the capabilities of intelligent geospatial data interpretation. These models combine various data modalities, such as optical, radar, and LiDAR imagery, with textual and geographic information, enabling more comprehensive analysis and understanding of remote sensing data. The integration of multiple modalities allows for improved performance in tasks like object detection, land cover classification, and change detection, which are often challenged by the complex and heterogeneous nature of remote sensing data. However, despite these advancements, several challenges remain. The diversity in data types, the need for large-scale annotated datasets, and the complexity of multimodal fusion techniques pose significant obstacles to the effective deployment of these models. Moreover, the computational demands of training and fine-tuning multimodal models require significant resources, further complicating their practical application in remote sensing image interpretation tasks. This paper provides a comprehensive review of the state-of-the-art in vision and multimodal foundation models for remote sensing, focusing on their architecture, training methods, datasets and application scenarios. We discuss the key challenges these models face, such as data alignment, cross-modal transfer learning, and scalability, while also identifying emerging research directions aimed at overcoming these limitations. Our goal is to provide a clear understanding of the current landscape of remote sensing foundation models and inspire future research that can push the boundaries of what these models can achieve in real-world applications. The list of resources collected by the paper can be found in the https://github.com/IRIP-BUAA/A-Review-for-remote-sensing-vision-language-models.
Land cover classification involves the production of land cover maps, which determine the type of land through remote sensing imagery. Over recent years, such classification is being performed by machine learning classification models, which can give highly accurate predictions on land cover per pixel using large quantities of input training data. However, such models do not currently take account of input measurement uncertainty, which is vital for traceability in metrology. In this work we propose a Bayesian classification framework using generative modelling to take account of input measurement uncertainty. We take the specific case of Bayesian quadratic discriminant analysis, and apply it to land cover datasets from Copernicus Sentinel-2 in 2020 and 2021. We benchmark the performance of the model against more popular classification models used in land cover maps such as random forests and neural networks. We find that such Bayesian models are more trustworthy, in the sense that they are more interpretable, explicitly model the input measurement uncertainty, and maintain predictive performance of class probability outputs across datasets of different years and sizes, whilst also being computationally efficient.
The Cerrado faces increasing environmental pressures, necessitating accurate land use and land cover (LULC) mapping despite challenges such as class imbalance and visually similar categories. To address this, we present CerraData-4MM, a multimodal dataset combining Sentinel-1 Synthetic Aperture Radar (SAR) and Sentinel-2 MultiSpectral Imagery (MSI) with 10m spatial resolution. The dataset includes two hierarchical classification levels with 7 and 14 classes, respectively, focusing on the diverse Bico do Papagaio ecoregion. We highlight CerraData-4MM's capacity to benchmark advanced semantic segmentation techniques by evaluating a standard U-Net and a more sophisticated Vision Transformer (ViT) model. The ViT achieves superior performance in multimodal scenarios, with the highest macro F1-score of 57.60% and a mean Intersection over Union (mIoU) of 49.05% at the first hierarchical level. Both models struggle with minority classes, particularly at the second hierarchical level, where U-Net's performance drops to an F1-score of 18.16%. Class balancing improves representation for underrepresented classes but reduces overall accuracy, underscoring the trade-off in weighted training. CerraData-4MM offers a challenging benchmark for advancing deep learning models to handle class imbalance and multimodal data fusion. Code, trained models, and data are publicly available at https://github.com/ai4luc/CerraData-4MM.
Synthetic Aperture Radar (SAR) images have proven to be a valuable cue for multimodal Land Cover Classification (LCC) when combined with RGB images. Most existing studies on cross-modal fusion assume that consistent feature information is necessary between the two modalities, and as a result, they construct networks without adequately addressing the unique characteristics of each modality. In this paper, we propose a novel architecture, named the Asymmetric Semantic Aligning Network (ASANet), which introduces asymmetry at the feature level to address the issue that multi-modal architectures frequently fail to fully utilize complementary features. The core of this network is the Semantic Focusing Module (SFM), which explicitly calculates differential weights for each modality to account for the modality-specific features. Furthermore, ASANet incorporates a Cascade Fusion Module (CFM), which delves deeper into channel and spatial representations to efficiently select features from the two modalities for fusion. Through the collaborative effort of these two modules, the proposed ASANet effectively learns feature correlations between the two modalities and eliminates noise caused by feature differences. Comprehensive experiments demonstrate that ASANet achieves excellent performance on three multimodal datasets. Additionally, we have established a new RGB-SAR multimodal dataset, on which our ASANet outperforms other mainstream methods with improvements ranging from 1.21% to 17.69%. The ASANet runs at 48.7 frames per second (FPS) when the input image is 256x256 pixels. The source code are available at https://github.com/whu-pzhang/ASANet
Recent self-supervised learning (SSL) methods have demonstrated impressive results in learning visual representations from unlabeled remote sensing images. However, most remote sensing images predominantly consist of scenographic scenes containing multiple ground objects without explicit foreground targets, which limits the performance of existing SSL methods that focus on foreground targets. This raises the question: Is there a method that can automatically aggregate similar objects within scenographic remote sensing images, thereby enabling models to differentiate knowledge embedded in various geospatial patterns for improved feature representation? In this work, we present the Pattern Integration and Enhancement Vision Transformer (PIEViT), a novel self-supervised learning framework designed specifically for remote sensing imagery. PIEViT utilizes a teacher-student architecture to address both image-level and patch-level tasks. It employs the Geospatial Pattern Cohesion (GPC) module to explore the natural clustering of patches, enhancing the differentiation of individual features. The Feature Integration Projection (FIP) module further refines masked token reconstruction using geospatially clustered patches. We validated PIEViT across multiple downstream tasks, including object detection, semantic segmentation, and change detection. Experiments demonstrated that PIEViT enhances the representation of internal patch features, providing significant improvements over existing self-supervised baselines. It achieves excellent results in object detection, land cover classification, and change detection, underscoring its robustness, generalization, and transferability for remote sensing image interpretation tasks.
Earth Observation (EO) data analysis has been significantly revolutionized by deep learning (DL), with applications typically limited to grid-like data structures. Graph Neural Networks (GNNs) emerge as an important innovation, propelling DL into the non-Euclidean domain. Naturally, GNNs can effectively tackle the challenges posed by diverse modalities, multiple sensors, and the heterogeneous nature of EO data. To introduce GNNs in the related domains, our review begins by offering fundamental knowledge on GNNs. Then, we summarize the generic problems in EO, to which GNNs can offer potential solutions. Following this, we explore a broad spectrum of GNNs' applications to scientific problems in Earth systems, covering areas such as weather and climate analysis, disaster management, air quality monitoring, agriculture, land cover classification, hydrological process modeling, and urban modeling. The rationale behind adopting GNNs in these fields is explained, alongside methodologies for organizing graphs and designing favorable architectures for various tasks. Furthermore, we highlight methodological challenges of implementing GNNs in these domains and possible solutions that could guide future research. While acknowledging that GNNs are not a universal solution, we conclude the paper by comparing them with other popular architectures like transformers and analyzing their potential synergies.
The segmentation of ultra-high resolution images poses challenges such as loss of spatial information or computational inefficiency. In this work, a novel approach that combines encoder-decoder architectures with domain decomposition strategies to address these challenges is proposed. Specifically, a domain decomposition-based U-Net (DDU-Net) architecture is introduced, which partitions input images into non-overlapping patches that can be processed independently on separate devices. A communication network is added to facilitate inter-patch information exchange to enhance the understanding of spatial context. Experimental validation is performed on a synthetic dataset that is designed to measure the effectiveness of the communication network. Then, the performance is tested on the DeepGlobe land cover classification dataset as a real-world benchmark data set. The results demonstrate that the approach, which includes inter-patch communication for images divided into $16\times16$ non-overlapping subimages, achieves a $2-3\,\%$ higher intersection over union (IoU) score compared to the same network without inter-patch communication. The performance of the network which includes communication is equivalent to that of a baseline U-Net trained on the full image, showing that our model provides an effective solution for segmenting ultra-high-resolution images while preserving spatial context. The code is available at https://github.com/corne00/DDU-Net.
When we are primarily interested in solving several problems jointly with a given prescribed high performance accuracy for each target application, then Foundation Models should for most cases be used rather than problem-specific models. We focus on the specific Computer Vision application of Foundation Models for Earth Observation (EO) and geospatial AI. These models can solve important problems we are tackling, including for example land cover classification, crop type mapping, flood segmentation, building density estimation, and road regression segmentation. In this paper, we show that for a limited number of labelled data, Foundation Models achieve improved performance compared to problem-specific models. In this work, we also present our proposed evaluation benchmark for Foundation Models for EO. Benchmarking the generalization performance of Foundation Models is important as it has become difficult to standardize a fair comparison across the many different models that have been proposed recently. We present the results using our evaluation benchmark for EO Foundation Models and show that Foundation Models are label efficient in the downstream tasks and help us solve problems we are tackling in EO and remote sensing.