Begüm Demir

Gencer Sumbul, Begüm Demir

The development of deep learning based image representation learning (IRL) methods has attracted great attention for various image understanding problems. Most of these methods require the availability of a high quantity and quality of annotated training images, which can be time-consuming and costly to gather. To reduce labeling costs, crowdsourced data, automatic labeling procedures or citizen science projects can be considered. However, such approaches increase the risk of including label noise in training data. It may result in overfitting on noisy labels when discriminative reasoning is employed. This leads to sub-optimal learning procedures, and thus inaccurate characterization of images. To address this, we introduce a generative reasoning integrated label noise robust deep representation learning (GRID) approach. Our approach aims to model the complementary characteristics of discriminative and generative reasoning for IRL under noisy labels. To this end, we first integrate generative reasoning into discriminative reasoning through a supervised variational autoencoder. This allows GRID to automatically detect training samples with noisy labels. Then, through our label noise robust hybrid representation learning strategy, GRID adjusts the whole learning procedure for IRL of these samples through generative reasoning and that of other samples through discriminative reasoning. Our approach learns discriminative image representations while preventing interference of noisy labels independently from the IRL method being selected. Thus, unlike the existing methods, GRID does not depend on the type of annotation, neural network architecture, loss function or learning task, and thus can be directly utilized for various problems. Experimental results show its effectiveness compared to state-of-the-art methods. The code of GRID is publicly available at https://github.com/gencersumbul/GRID.

Michael Mommert, Nicolas Kesseli, Joëlle Hanna et al. · berkeley

Deep learning methods have proven to be a powerful tool in the analysis of large amounts of complex Earth observation data. However, while Earth observation data are multi-modal in most cases, only single or few modalities are typically considered. In this work, we present the ben-ge dataset, which supplements the BigEarthNet-MM dataset by compiling freely and globally available geographical and environmental data. Based on this dataset, we showcase the value of combining different data modalities for the downstream tasks of patch-based land-use/land-cover classification and land-use/land-cover segmentation. ben-ge is freely available and expected to serve as a test bed for fully supervised and self-supervised Earth observation applications.

Georgii Mikriukov, Mahdyar Ravanbakhsh, Begüm Demir

The development of cross-modal retrieval systems that can search and retrieve semantically relevant data across different modalities based on a query in any modality has attracted great attention in remote sensing (RS). In this paper, we focus our attention on cross-modal text-image retrieval, where queries from one modality (e.g., text) can be matched to archive entries from another (e.g., image). Most of the existing cross-modal text-image retrieval systems in RS require a high number of labeled training samples and also do not allow fast and memory-efficient retrieval. These issues limit the applicability of the existing cross-modal retrieval systems for large-scale applications in RS. To address this problem, in this paper we introduce a novel unsupervised cross-modal contrastive hashing (DUCH) method for text-image retrieval in RS. To this end, the proposed DUCH is made up of two main modules: 1) feature extraction module, which extracts deep representations of two modalities; 2) hashing module that learns to generate cross-modal binary hash codes from the extracted representations. We introduce a novel multi-objective loss function including: i) contrastive objectives that enable similarity preservation in intra- and inter-modal similarities; ii) an adversarial objective that is enforced across two modalities for cross-modal representation consistency; and iii) binarization objectives for generating hash codes. Experimental results show that the proposed DUCH outperforms state-of-the-art methods. Our code is publicly available at https://git.tu-berlin.de/rsim/duch.

Martin Hermann Paul Fuchs, Begüm Demir

The development of learning-based hyperspectral image compression methods has recently attracted great attention in remote sensing. Such methods require a high number of hyperspectral images to be used during training to optimize all parameters and reach a high compression performance. However, existing hyperspectral datasets are not sufficient to train and evaluate learning-based compression methods, which hinders the research in this field. To address this problem, in this paper we present HySpecNet-11k that is a large-scale hyperspectral benchmark dataset made up of 11,483 nonoverlapping image patches. Each patch is a portion of 128 $\times$ 128 pixels with 224 spectral bands and a ground sample distance of 30 m. We exploit HySpecNet-11k to benchmark the current state of the art in learning-based hyperspectral image compression by focussing our attention on various 1D, 2D and 3D convolutional autoencoder architectures. Nevertheless, HySpecNet-11k can be used for any unsupervised learning task in the framework of hyperspectral image analysis. The dataset, our code and the pre-trained weights are publicly available at https://hyspecnet.rsim.berlin

Tim Siebert, Kai Norman Clasen, Mahdyar Ravanbakhsh et al.

With the new generation of satellite technologies, the archives of remote sensing (RS) images are growing very fast. To make the intrinsic information of each RS image easily accessible, visual question answering (VQA) has been introduced in RS. VQA allows a user to formulate a free-form question concerning the content of RS images to extract generic information. It has been shown that the fusion of the input modalities (i.e., image and text) is crucial for the performance of VQA systems. Most of the current fusion approaches use modality-specific representations in their fusion modules instead of joint representation learning. However, to discover the underlying relation between both the image and question modality, the model is required to learn the joint representation instead of simply combining (e.g., concatenating, adding, or multiplying) the modality-specific representations. We propose a multi-modal transformer-based architecture to overcome this issue. Our proposed architecture consists of three main modules: i) the feature extraction module for extracting the modality-specific features; ii) the fusion module, which leverages a user-defined number of multi-modal transformer layers of the VisualBERT model (VB); and iii) the classification module to obtain the answer. Experimental results obtained on the RSVQAxBEN and RSVQA-LR datasets (which are made up of RGB bands of Sentinel-2 images) demonstrate the effectiveness of VBFusion for VQA tasks in RS. To analyze the importance of using other spectral bands for the description of the complex content of RS images in the framework of VQA, we extend the RSVQAxBEN dataset to include all the spectral bands of Sentinel-2 images with 10m and 20m spatial resolution.

Tom Burgert, Mahdyar Ravanbakhsh, Begüm Demir

The development of accurate methods for multi-label classification (MLC) of remote sensing (RS) images is one of the most important research topics in RS. To address MLC problems, the use of deep neural networks that require a high number of reliable training images annotated by multiple land-cover class labels (multi-labels) has been found popular in RS. However, collecting such annotations is time-consuming and costly. A common procedure to obtain annotations at zero labeling cost is to rely on thematic products or crowdsourced labels. As a drawback, these procedures come with the risk of label noise that can distort the learning process of the MLC algorithms. In the literature, most label noise robust methods are designed for single-label classification (SLC) problems in computer vision (CV), where each image is annotated by a single label. Unlike SLC, label noise in MLC can be associated with: 1) subtractive label-noise (a land cover class label is not assigned to an image while that class is present in the image); 2) additive label-noise (a land cover class label is assigned to an image although that class is not present in the given image); and 3) mixed label-noise (a combination of both). In this paper, we investigate three different noise robust CV SLC methods and adapt them to be robust for multi-label noise scenarios in RS. During experiments, we study the effects of different types of multi-label noise and evaluate the adapted methods rigorously. To this end, we also introduce a synthetic multi-label noise injection strategy that is more adequate to simulate operational scenarios compared to the uniform label noise injection strategy, in which the labels of absent and present classes are flipped at uniform probability. Further, we study the relevance of different evaluation metrics in MLC problems under noisy multi-labels.

Ahmet Kerem Aksoy, Pavel Dushev, Eleni Tzirita Zacharatou et al.

The growing operational capability of global Earth Observation (EO) creates new opportunities for data-driven approaches to understand and protect our planet. However, the current use of EO archives is very restricted due to the huge archive sizes and the limited exploration capabilities provided by EO platforms. To address this limitation, we have recently proposed MiLaN, a content-based image retrieval approach for fast similarity search in satellite image archives. MiLaN is a deep hashing network based on metric learning that encodes high-dimensional image features into compact binary hash codes. We use these codes as keys in a hash table to enable real-time nearest neighbor search and highly accurate retrieval. In this demonstration, we showcase the efficiency of MiLaN by integrating it with EarthQube, a browser and search engine within AgoraEO. EarthQube supports interactive visual exploration and Query-by-Example over satellite image repositories. Demo visitors will interact with EarthQube playing the role of different users that search images in a large-scale remote sensing archive by their semantic content and apply other filters.

Barış Büyüktaş, Gencer Sumbul, Begüm Demir

The development of federated learning (FL) methods, which aim to learn from distributed databases (i.e., clients) without accessing data on clients, has recently attracted great attention. Most of these methods assume that the clients are associated with the same data modality. However, remote sensing (RS) images in different clients can be associated with different data modalities that can improve the classification performance when jointly used. To address this problem, in this paper we introduce a novel multi-modal FL framework that aims to learn from decentralized multi-modal RS image archives for RS image classification problems. The proposed framework is made up of three modules: 1) multi-modal fusion (MF); 2) feature whitening (FW); and 3) mutual information maximization (MIM). The MF module performs iterative model averaging to learn without accessing data on clients in the case that clients are associated with different data modalities. The FW module aligns the representations learned among the different clients. The MIM module maximizes the similarity of images from different modalities. Experimental results show the effectiveness of the proposed framework compared to iterative model averaging, which is a widely used algorithm in FL. The code of the proposed framework is publicly available at https://git.tu-berlin.de/rsim/MM-FL.

Barış Büyüktaş, Gencer Sumbul, Begüm Demir

Federated learning (FL) enables the collaboration of multiple deep learning models to learn from decentralized data archives (i.e., clients) without accessing data on clients. Although FL offers ample opportunities in knowledge discovery from distributed image archives, it is seldom considered in remote sensing (RS). In this paper, as a first time in RS, we present a comparative study of state-of-the-art FL algorithms for RS image classification problems. To this end, we initially provide a systematic review of the FL algorithms presented in the computer vision and machine learning communities. Then, we select several state-of-the-art FL algorithms based on their effectiveness with respect to training data heterogeneity across clients (known as non-IID data). After presenting an extensive overview of the selected algorithms, a theoretical comparison of the algorithms is conducted based on their: 1) local training complexity; 2) aggregation complexity; 3) learning efficiency; 4) communication cost; and 5) scalability in terms of number of clients. After the theoretical comparison, experimental analyses are presented to compare them under different decentralization scenarios. For the experimental analyses, we focus our attention on multi-label image classification problems in RS. Based on our comprehensive analyses, we finally derive a guideline for selecting suitable FL algorithms in RS. The code of this work is publicly available at https://git.tu-berlin.de/rsim/FL-RS.

David Hoffmann, Kai Norman Clasen, Begüm Demir

In this paper, we introduce a novel Synchronized Class Token Fusion (SCT Fusion) architecture in the framework of multi-modal multi-label classification (MLC) of remote sensing (RS) images. The proposed architecture leverages modality-specific attention-based transformer encoders to process varying input modalities, while exchanging information across modalities by synchronizing the special class tokens after each transformer encoder block. The synchronization involves fusing the class tokens with a trainable fusion transformation, resulting in a synchronized class token that contains information from all modalities. As the fusion transformation is trainable, it allows to reach an accurate representation of the shared features among different modalities. Experimental results show the effectiveness of the proposed architecture over single-modality architectures and an early fusion multi-modal architecture when evaluated on a multi-modal MLC dataset. The code of the proposed architecture is publicly available at https://git.tu-berlin.de/rsim/sct-fusion.

Soronzonbold Otgonbaatar, Mihai Datcu, Begüm Demir

Machine Learning (ML) techniques are employed to analyze and process big Remote Sensing (RS) data, and one well-known ML technique is a Support Vector Machine (SVM). An SVM is a quadratic programming (QP) problem, and a D-Wave quantum annealer (D-Wave QA) promises to solve this QP problem more efficiently than a conventional computer. However, the D-Wave QA cannot solve directly the SVM due to its very few input qubits. Hence, we use a coreset ("core of a dataset") of given EO data for training an SVM on this small D-Wave QA. The coreset is a small, representative weighted subset of an original dataset, and any training models generate competitive classes by using the coreset in contrast to by using its original dataset. We measured the closeness between an original dataset and its coreset by employing a Kullback-Leibler (KL) divergence measure. Moreover, we trained the SVM on the coreset data by using both a D-Wave QA and a conventional method. We conclude that the coreset characterizes the original dataset with very small KL divergence measure. In addition, we present our KL divergence results for demonstrating the closeness between our original data and its coreset. As practical RS data, we use Hyperspectral Image (HSI) of Indian Pine, USA.

Lars Möllenbrok, Gencer Sumbul, Begüm Demir

In this letter, we introduce deep active learning (AL) for multi-label classification (MLC) problems in remote sensing (RS). In particular, we investigate the effectiveness of several AL query functions for MLC of RS images. Unlike the existing AL query functions (which are defined for single-label classification or semantic segmentation problems), each query function in this paper is based on the evaluation of two criteria: i) multi-label uncertainty; and ii) multi-label diversity. The multi-label uncertainty criterion is associated to the confidence of the deep neural networks (DNNs) in correctly assigning multi-labels to each image. To assess this criterion, we investigate three strategies: i) learning multi-label loss ordering; ii) measuring temporal discrepancy of multi-label predictions; and iii) measuring magnitude of approximated gradient embeddings. The multi-label diversity criterion is associated to the selection of a set of images that are as diverse as possible to each other that prevents redundancy among them. To assess this criterion, we exploit a clustering based strategy. We combine each of the above-mentioned uncertainty strategies with the clustering based diversity strategy, resulting in three different query functions. All the considered query functions are introduced for the first time in the framework of MLC problems in RS. Experimental results obtained on two benchmark archives show that these query functions result in the selection of a highly informative set of samples at each iteration of the AL process.

Kai Norman Clasen, Leonard Hackel, Tom Burgert et al.

This paper presents refined BigEarthNet (reBEN) that is a large-scale, multi-modal remote sensing dataset constructed to support deep learning (DL) studies for remote sensing image analysis. The reBEN dataset consists of 549,488 pairs of Sentinel-1 and Sentinel-2 image patches. To construct reBEN, we initially consider the Sentinel-1 and Sentinel-2 tiles used to construct the BigEarthNet dataset and then divide them into patches of size 1200 m x 1200 m. We apply atmospheric correction to the Sentinel-2 patches using the latest version of the sen2cor tool, resulting in higher-quality patches compared to those present in BigEarthNet. Each patch is then associated with a pixel-level reference map and scene-level multi-labels. This makes reBEN suitable for pixel- and scene-based learning tasks. The labels are derived from the most recent CORINE Land Cover (CLC) map of 2018 by utilizing the 19-class nomenclature as in BigEarthNet. The use of the most recent CLC map results in overcoming the label noise present in BigEarthNet. Furthermore, we introduce a new geographical-based split assignment algorithm that significantly reduces the spatial correlation among the train, validation, and test sets with respect to those present in BigEarthNet. This increases the reliability of the evaluation of DL models. To minimize the DL model training time, we introduce software tools that convert the reBEN dataset into a DL-optimized data format. In our experiments, we show the potential of reBEN for multi-modal multi-label image classification problems by considering several state-of-the-art DL models. The pre-trained model weights, associated code, and complete dataset are available at https://bigearth.net.

Julia Henkel, Genc Hoxha, Gencer Sumbul et al.

Deep metric learning (DML) based methods have been found very effective for content-based image retrieval (CBIR) in remote sensing (RS). For accurately learning the model parameters of deep neural networks, most of the DML methods require a high number of annotated training images, which can be costly to gather. To address this problem, in this paper we present an annotation cost efficient active learning (AL) method (denoted as ANNEAL). The proposed method aims to iteratively enrich the training set by annotating the most informative image pairs as similar or dissimilar, while accurately modelling a deep metric space. This is achieved by two consecutive steps. In the first step the pairwise image similarity is modelled based on the available training set. Then, in the second step the most uncertain and diverse (i.e., informative) image pairs are selected to be annotated. Unlike the existing AL methods for CBIR, at each AL iteration of ANNEAL a human expert is asked to annotate the most informative image pairs as similar/dissimilar. This significantly reduces the annotation cost compared to annotating images with land-use/land cover class labels. Experimental results show the effectiveness of our method. The code of ANNEAL is publicly available at https://git.tu-berlin.de/rsim/ANNEAL.

Lars Möllenbrok, Begüm Demir

In recent years, deep neural networks (DNNs) have been found very successful for multi-label classification (MLC) of remote sensing (RS) images. Self-supervised pre-training combined with fine-tuning on a randomly selected small training set has become a popular approach to minimize annotation efforts of data-demanding DNNs. However, fine-tuning on a small and biased training set may limit model performance. To address this issue, we investigate the effectiveness of the joint use of self-supervised pre-training with active learning (AL). The considered AL strategy aims at guiding the MLC fine-tuning of a self-supervised model by selecting informative training samples to annotate in an iterative manner. Experimental results show the effectiveness of applying AL-guided fine-tuning (particularly for the case where strong class-imbalance is present in MLC problems) compared to the application of fine-tuning using a randomly constructed small training set.

Tom Burgert, Leonard Hackel, Paolo Rota et al.

Self-supervised learning (SSL) has become a powerful paradigm for learning from large, unlabeled datasets, particularly in computer vision (CV). However, applying SSL to multispectral remote sensing (RS) images presents unique challenges and opportunities due to the geographical and temporal variability of the data. In this paper, we introduce GeoRank, a novel regularization method for contrastive SSL that improves upon prior techniques by directly optimizing spherical distances to embed geographical relationships into the learned feature space. GeoRank outperforms or matches prior methods that integrate geographical metadata and consistently improves diverse contrastive SSL algorithms (e.g., BYOL, DINO). Beyond this, we present a systematic investigation of key adaptations of contrastive SSL for multispectral RS images, including the effectiveness of data augmentations, the impact of dataset cardinality and image size on performance, and the task dependency of temporal views. Code is available at https://github.com/tomburgert/georank.

Gencer Sumbul, Begüm Demir

Due to the publicly available thematic maps and crowd-sourced data, remote sensing (RS) image annotations can be gathered at zero cost for training deep neural networks (DNNs). However, such annotation sources may increase the risk of including noisy labels in training data, leading to inaccurate RS image representation learning (IRL). To address this issue, in this paper we propose a label noise robust IRL method that aims to prevent the interference of noisy labels on IRL, independently from the learning task being considered in RS. To this end, the proposed method combines a supervised variational autoencoder (SVAE) with any kind of DNN. This is achieved by defining variational generative process based on image features. This allows us to define the importance of each training sample for IRL based on the loss values acquired from the SVAE and the task head of the considered DNN. Then, the proposed method imposes lower importance to images with noisy labels, while giving higher importance to those with correct labels during IRL. Experimental results show the effectiveness of the proposed method when compared to well-known label noise robust IRL methods applied to RS images. The code of the proposed method is publicly available at https://git.tu-berlin.de/rsim/RS-IRL-SVAE.

Leonard Hackel, Kai Norman Clasen, Mahdyar Ravanbakhsh et al.

Visual question answering (VQA) methods in remote sensing (RS) aim to answer natural language questions with respect to an RS image. Most of the existing methods require a large amount of computational resources, which limits their application in operational scenarios in RS. To address this issue, in this paper we present an effective lightweight transformer-based VQA in RS (LiT-4-RSVQA) architecture for efficient and accurate VQA in RS. Our architecture consists of: i) a lightweight text encoder module; ii) a lightweight image encoder module; iii) a fusion module; and iv) a classification module. The experimental results obtained on a VQA benchmark dataset demonstrate that our proposed LiT-4-RSVQA architecture provides accurate VQA results while significantly reducing the computational requirements on the executing hardware. Our code is publicly available at https://git.tu-berlin.de/rsim/lit4rsvqa.

Martin Hermann Paul Fuchs, Behnood Rasti, Begüm Demir

The development of learning-based hyperspectral image (HSI) compression models has recently attracted significant interest. Existing models predominantly utilize convolutional filters, which capture only local dependencies. Furthermore,they often incur high training costs and exhibit substantial computational complexity. To address these limitations, in this paper we propose Hyperspectral Compression Transformer (HyCoT) that is a transformer-based autoencoder for pixelwise HSI compression. Additionally, we apply a simple yet effective training set reduction approach to accelerate the training process. Experimental results on the HySpecNet-11k dataset demonstrate that HyCoT surpasses the state of the art across various compression ratios by over 1 dB of PSNR with significantly reduced computational requirements. Our code and pre-trained weights are publicly available at https://git.tu-berlin.de/rsim/hycot .

Tom-Lukas Breitkopf, Leonard W. Hackel, Mahdyar Ravanbakhsh et al.

Subsurface tile drainage pipes provide agronomic, economic and environmental benefits. By lowering the water table of wet soils, they improve the aeration of plant roots and ultimately increase the productivity of farmland. They do however also provide an entryway of agrochemicals into subsurface water bodies and increase nutrition loss in soils. For maintenance and infrastructural development, accurate maps of tile drainage pipe locations and drained agricultural land are needed. However, these maps are often outdated or not present. Different remote sensing (RS) image processing techniques have been applied over the years with varying degrees of success to overcome these restrictions. Recent developments in deep learning (DL) techniques improve upon the conventional techniques with machine learning segmentation models. In this study, we introduce two DL-based models: i) improved U-Net architecture; and ii) Visual Transformer-based encoder-decoder in the framework of tile drainage pipe detection. Experimental results confirm the effectiveness of both models in terms of detection accuracy when compared to a basic U-Net architecture. Our code and models are publicly available at https://git.tu-berlin.de/rsim/drainage-pipes-detection.

Angelos Zavras, Dimitrios Michail, Begüm Demir et al.

Deep Learning (DL) is undergoing a paradigm shift with the emergence of foundation models. In this work, we focus on Contrastive Language-Image Pre-training (CLIP), a Vision-Language foundation model that achieves high accuracy across various image classification tasks and often rivals fully supervised baselines, despite not being explicitly trained for those tasks. Nevertheless, there are still domains where zero-shot CLIP performance is far from optimal, such as Remote Sensing (RS) and medical imagery. These domains do not only exhibit fundamentally different distributions compared to natural images, but also commonly rely on complementary modalities, beyond RGB, to derive meaningful insights. To this end, we propose a methodology to align distinct RS image modalities with the visual and textual modalities of CLIP. Our two-stage procedure addresses the aforementioned distribution shift, extends the zero-shot capabilities of CLIP and enriches CLIP's shared embedding space with domain-specific knowledge. Initially, we robustly fine-tune CLIP according to the PAINT (Ilharco et al., 2022) patching protocol, in order to deal with the distribution shift. Building upon this foundation, we facilitate the cross-modal alignment of a RS modality encoder by distilling knowledge from the CLIP visual and textual encoders. We empirically show that both patching and cross-modal alignment translate to significant performance gains, across several RS imagery classification and cross-modal retrieval benchmark datasets. Notably, these enhancements are achieved without the reliance on textual descriptions, without introducing any task-specific parameters, without training from scratch and without catastrophic forgetting. We make our code implementation and weights for all experiments publicly available at https://github.com/Orion-AI-Lab/MindTheModalityGap.

Jakob Hackstein, Gencer Sumbul, Kai Norman Clasen et al.

Self-supervised learning through masked autoencoders (MAEs) has recently attracted great attention for remote sensing (RS) image representation learning, and thus embodies a significant potential for content-based image retrieval (CBIR) from ever-growing RS image archives. However, the existing MAE based CBIR studies in RS assume that the considered RS images are acquired by a single image sensor, and thus are only suitable for uni-modal CBIR problems. The effectiveness of MAEs for cross-sensor CBIR, which aims to search semantically similar images across different image modalities, has not been explored yet. In this paper, we take the first step to explore the effectiveness of MAEs for sensor-agnostic CBIR in RS. To this end, we present a systematic overview on the possible adaptations of the vanilla MAE to exploit masked image modeling on multi-sensor RS image archives (denoted as cross-sensor masked autoencoders [CSMAEs]) in the context of CBIR. Based on different adjustments applied to the vanilla MAE, we introduce different CSMAE models. We also provide an extensive experimental analysis of these CSMAE models. We finally derive a guideline to exploit masked image modeling for uni-modal and cross-modal CBIR problems in RS. The code of this work is publicly available at https://github.com/jakhac/CSMAE.

Panagiotis Agrafiotis, Begüm Demir

Accurate, detailed, and high-frequent bathymetry is crucial for shallow seabed areas facing intense climatological and anthropogenic pressures. Current methods utilizing airborne or satellite optical imagery to derive bathymetry primarily rely on either SfM-MVS with refraction correction or Spectrally Derived Bathymetry (SDB). However, SDB methods often require extensive manual fieldwork or costly reference data, while SfM-MVS approaches face challenges even after refraction correction. These include depth data gaps and noise in environments with homogeneous visual textures, which hinder the creation of accurate and complete Digital Surface Models (DSMs) of the seabed. To address these challenges, this work introduces a methodology that combines the high-fidelity 3D reconstruction capabilities of the SfM-MVS methods with state-of-the-art refraction correction techniques, along with the spectral analysis capabilities of a new deep learning-based method for bathymetry prediction. This integration enables a synergistic approach where SfM-MVS derived DSMs with data gaps are used as training data to generate complete bathymetric maps. In this context, we propose Swin-BathyUNet that combines U-Net with Swin Transformer self-attention layers and a cross-attention mechanism, specifically tailored for SDB. Swin-BathyUNet is designed to improve bathymetric accuracy by capturing long-range spatial relationships and can also function as a standalone solution for standard SDB with various training depth data, independent of the SfM-MVS output. Experimental results in two completely different test sites in the Mediterranean and Baltic Seas demonstrate the effectiveness of the proposed approach through extensive experiments that demonstrate improvements in bathymetric accuracy, detail, coverage, and noise reduction in the predicted DSM. The code is available at https://github.com/pagraf/Swin-BathyUNet.

Tom Burgert, Kai Norman Clasen, Jonas Klotz et al.

The development of supervised deep learning-based methods for multi-label scene classification (MLC) is one of the prominent research directions in remote sensing (RS). However, collecting annotations for large RS image archives is time-consuming and costly. To address this issue, several data augmentation methods have been introduced in RS. Among others, the CutMix data augmentation technique, which combines parts of two existing training images to generate an augmented image, stands out as a particularly effective approach. However, the direct application of CutMix in RS MLC can lead to the erasure or addition of class labels (i.e., label noise) in the augmented (i.e., combined) training image. To address this problem, we introduce a label propagation (LP) strategy that allows the effective application of CutMix in the context of MLC problems in RS without being affected by label noise. To this end, our proposed LP strategy exploits pixel-level class positional information to update the multi-label of the augmented training image. We propose to access such class positional information from reference maps (e.g., thematic products) associated with each training image or from class explanation masks provided by an explanation method if no reference maps are available. Similarly to pairing two training images, our LP strategy carries out a pairing operation on the associated pixel-level class positional information to derive the updated multi-label for the augmented image. Experimental results show the effectiveness of our LP strategy in general (e.g., an improvement of 2% to 4% mAP macro compared to standard CutMix) and its robustness in the case of various simulated and real scenarios with noisy class positional information in particular. Code is available at https://git.tu-berlin.de/rsim/cutmix_lp.

Tom Burgert, Julia Henkel, Begüm Demir

The development of reliable methods for multi-label classification (MLC) has become a prominent research direction in remote sensing (RS). As the scale of RS data continues to expand, annotation procedures increasingly rely on thematic products or crowdsourced procedures to reduce the cost of manual annotation. While cost-effective, these strategies often introduce multi-label noise in the form of partially incorrect annotations. In MLC, label noise arises as additive noise, subtractive noise, or a combination of both in the form of mixed noise. Previous work has largely overlooked this distinction and commonly treats noisy annotations as supervised signals, lacking mechanisms that explicitly adapt learning behavior to different noise types. To address this limitation, we propose NAR, a noise-adaptive regularization method that explicitly distinguishes between additive and subtractive noise within a semi-supervised learning framework. NAR employs a confidence-based label handling mechanism that dynamically retains label entries with high confidence, temporarily deactivates entries with moderate confidence, and corrects low confidence entries via flipping. This selective attenuation of supervision is integrated with early-learning regularization (ELR) to stabilize training and mitigate overfitting to corrupted labels. Experiments across additive, subtractive, and mixed noise scenarios demonstrate that NAR consistently improves robustness compared with existing methods. Performance improvements are most pronounced under subtractive and mixed noise, indicating that adaptive suppression and selective correction of noisy supervision provide an effective strategy for noise robust learning in RS MLC.

Leonard Hackel, Tom Burgert, Begüm Demir

Large-scale foundation models (FMs) in remote sensing (RS) are developed based on the paradigms established in computer vision (CV) and have shown promise for various Earth observation applications. However, the direct transfer of scaling assumptions from CV to RS has not been adequately examined. We hypothesize that RS FMs enter an overparameterized regime at substantially smaller scales than their CV counterparts, where increasing parameter count primarily induces redundant representations rather than qualitatively new abstractions. To test this hypothesis, we use post-hoc slimming, where we uniformly reduce the width of pretrained encoder, as a tool to measure representational redundancy across six state-of-the-art RS FMs on four downstream classification tasks. Our findings reveal a significant contrast with those in the CV domain: while a post-hoc slimmed masked autoencoder (MAE) trained on ImageNet retains less than 10% accuracy at 1% FLOPs, RS FMs maintain over 71% relative accuracy at the same budget. This sevenfold difference provides strong empirical support for our hypothesis. We further demonstrate that learned slimmable training can improve both Momentum Contrast (MoCo)- and MAE- based models. In addition, through the explained variance ratio and the feature correlation analysis, we provide mechanistic explanations showing that RS FMs distribute task-relevant information with high redundancy. Our findings establish post-hoc slimmability as both a practical deployment strategy for resource-constrained environments and a diagnostic tool that challenges the prevailing scaling paradigm in RS. Upon acceptance, we will publish all code.

Panagiotis Agrafiotis, Begüm Demir

Accurate, detailed, and regularly updated bathymetry, coupled with complex semantic content, is essential for under-mapped shallow-water environments facing increasing climatological and anthropogenic pressures. However, existing approaches that derive either depth or seabed classes from remote sensing imagery treat these tasks in isolation, forfeiting the mutual benefits of their interaction and hindering the broader adoption of deep learning methods. To address these limitations, we introduce Seabed-Net, a unified multi-task framework that simultaneously predicts bathymetry and pixel-based seabed classification from remote sensing imagery of various resolutions. Seabed-Net employs dual-branch encoders for bathymetry estimation and pixel-based seabed classification, integrates cross-task features via an Attention Feature Fusion module and a windowed Swin-Transformer fusion block, and balances objectives through dynamic task uncertainty weighting. In extensive evaluations at two heterogeneous coastal sites, it consistently outperforms traditional empirical models and traditional machine learning regression methods, achieving up to 75\% lower RMSE. It also reduces bathymetric RMSE by 10-30\% compared to state-of-the-art single-task and multi-task baselines and improves seabed classification accuracy up to 8\%. Qualitative analyses further demonstrate enhanced spatial consistency, sharper habitat boundaries, and corrected depth biases in low-contrast regions. These results confirm that jointly modeling depth with both substrate and seabed habitats yields synergistic gains, offering a robust, open solution for integrated shallow-water mapping. Code and pretrained weights are available at https://github.com/pagraf/Seabed-Net.

Tom Burgert, Oliver Stoll, Paolo Rota et al.

The hypothesis that Convolutional Neural Networks (CNNs) are inherently texture-biased has shaped much of the discourse on feature use in deep learning. We revisit this hypothesis by examining limitations in the cue-conflict experiment by Geirhos et al. To address these limitations, we propose a domain-agnostic framework that quantifies feature reliance through systematic suppression of shape, texture, and color cues, avoiding the confounds of forced-choice conflicts. By evaluating humans and neural networks under controlled suppression conditions, we find that CNNs are not inherently texture-biased but predominantly rely on local shape features. Nonetheless, this reliance can be substantially mitigated through modern training strategies or architectures (ConvNeXt, ViTs). We further extend the analysis across computer vision, medical imaging, and remote sensing, revealing that reliance patterns differ systematically: computer vision models prioritize shape, medical imaging models emphasize color, and remote sensing models exhibit a stronger reliance on texture. Code is available at https://github.com/tomburgert/feature-reliance.

Martin Hermann Paul Fuchs, Behnood Rasti, Begüm Demir

With the rapid growth of hyperspectral data archives in remote sensing (RS), the need for efficient storage has become essential, driving significant attention toward learning-based hyperspectral image (HSI) compression. However, a comprehensive investigation of the individual and joint effects of spectral and spatial compression on learning-based HSI compression has not been thoroughly examined yet. Conducting such an analysis is crucial for understanding how the exploitation of spectral, spatial, and joint spatio-spectral redundancies affects HSI compression. To address this issue, we propose Adjustable Spatio-Spectral Hyperspectral Image Compression Network (HyCASS), a learning-based model designed for adjustable HSI compression in both spectral and spatial dimensions. HyCASS consists of six main modules: 1) spectral encoder module; 2) spatial encoder module; 3) compression ratio (CR) adapter encoder module; 4) CR adapter decoder module; 5) spatial decoder module; and 6) spectral decoder module. The modules employ convolutional layers and transformer blocks to capture both short-range and long-range redundancies. Experimental results on three HSI benchmark datasets demonstrate the effectiveness of our proposed adjustable model compared to existing learning-based compression models, surpassing the state of the art by up to 2.36 dB in terms of PSNR. Based on our results, we establish a guideline for effectively balancing spectral and spatial compression across different CRs, taking into account the spatial resolution of the HSIs. Our code and pre-trained model weights are publicly available at https://git.tu-berlin.de/rsim/hycass .

Lars Möllenbrok, Behnood Rasti, Begüm Demir

The development of continual learning (CL) methods, which aim to learn new tasks in a sequential manner from the training data acquired continuously, has gained great attention in remote sensing (RS). The existing CL methods in RS, while learning new tasks, enhance robustness towards catastrophic forgetting. This is achieved by using a large number of labeled training samples, which is costly and not always feasible to gather in RS. To address this problem, we propose a novel continual self-supervised learning method in the context of masked autoencoders (denoted as CoSMAE). The proposed CoSMAE consists of two components: i) data mixup; and ii) model mixup knowledge distillation. Data mixup is associated with retaining information on previous data distributions by interpolating images from the current task with those from the previous tasks. Model mixup knowledge distillation is associated with distilling knowledge from past models and the current model simultaneously by interpolating their model weights to form a teacher for the knowledge distillation. The two components complement each other to regularize the MAE at the data and model levels to facilitate better generalization across tasks and reduce the risk of catastrophic forgetting. Experimental results show that CoSMAE achieves significant improvements of up to 4.94% over state-of-the-art CL methods applied to MAE. Our code is publicly available at: https://git.tu-berlin.de/rsim/CoSMAE.

Barış Büyüktaş, Kenneth Weitzel, Sebastian Völkers et al.

Federated learning (FL) aims to collaboratively learn deep learning model parameters from decentralized data archives (i.e., clients) without accessing training data on clients. However, the training data across clients might be not independent and identically distributed (non-IID), which may result in difficulty in achieving optimal model convergence. In this work, we investigate the capability of state-of-the-art transformer architectures (which are MLP-Mixer, ConvMixer, PoolFormer) to address the challenges related to non-IID training data across various clients in the context of FL for multi-label classification (MLC) problems in remote sensing (RS). The considered transformer architectures are compared among themselves and with the ResNet-50 architecture in terms of their: 1) robustness to training data heterogeneity; 2) local training complexity; and 3) aggregation complexity under different non-IID levels. The experimental results obtained on the BigEarthNet-S2 benchmark archive demonstrate that the considered architectures increase the generalization ability with the cost of higher local training and aggregation complexities. On the basis of our analysis, some guidelines are derived for a proper selection of transformer architecture in the context of FL for RS MLC. The code of this work is publicly available at https://git.tu-berlin.de/rsim/FL-Transformer.

Fabian Jaensch, Giuseppe Caire, Begüm Demir

Over the last years, several works have explored the application of deep learning algorithms to determine the large-scale signal fading (also referred to as ``path loss'') between transmitter and receiver pairs in urban communication networks. The central idea is to replace costly measurement campaigns, inaccurate statistical models or computationally expensive ray-tracing simulations by machine learning models which, once trained, produce accurate predictions almost instantly. Although the topic has attracted attention from many researchers, there are few open benchmark datasets and codebases that would allow everyone to test and compare the developed methods and algorithms. We take a step towards filling this gap by releasing a publicly available dataset of simulated path loss radio maps together with realistic city maps from real-world locations and aerial images from open datasources. Initial experiments regarding model architectures, input feature design and estimation of radio maps from aerial images are presented and the code is made available.

Panagiotis Agrafiotis, Łukasz Janowski, Dimitrios Skarlatos et al.

Accurate, detailed, and high-frequent bathymetry, coupled with complex semantic content, is crucial for the undermapped shallow seabed areas facing intense climatological and anthropogenic pressures. Current methods exploiting remote sensing images to derive bathymetry or seabed classes mainly exploit non-open data. This lack of openly accessible benchmark archives prevents the wider use of deep learning methods in such applications. To address this issue, in this paper we present the MagicBathyNet, which is a benchmark dataset made up of image patches of Sentinel2, SPOT-6 and aerial imagery, bathymetry in raster format and annotations of seabed classes. MagicBathyNet is then exploited to benchmark state-of-the-art methods in learning-based bathymetry and pixel-based classification. Dataset, pre-trained weights, and code are publicly available at www.magicbathy.eu/magicbathynet.html.

Tom Burgert, Begüm Demir

The application of data augmentation for deep learning (DL) methods plays an important role in achieving state-of-the-art results in supervised, semi-supervised, and self-supervised image classification. In particular, channel transformations (e.g., solarize, grayscale, brightness adjustments) are integrated into data augmentation pipelines for remote sensing (RS) image classification tasks. However, contradicting beliefs exist about their proper applications to RS images. A common point of critique is that the application of channel augmentation techniques may lead to physically inconsistent spectral data (i.e., pixel signatures). To shed light on the open debate, we propose an approach to estimate whether a channel augmentation technique affects the physical information of RS images. To this end, the proposed approach estimates a score that measures the alignment of a pixel signature within a time series that can be naturally subject to deviations caused by factors such as acquisition conditions or phenological states of vegetation. We compare the scores associated with original and augmented pixel signatures to evaluate the physical consistency. Experimental results on a multi-label image classification task show that channel augmentations yielding a score that exceeds the expected deviation of original pixel signatures can not improve the performance of a baseline model trained without augmentation.

Angelos Zavras, Dimitrios Michail, Xiao Xiang Zhu et al.

The continuous operation of Earth-orbiting satellites generates vast and ever-growing archives of Remote Sensing (RS) images. Natural language presents an intuitive interface for accessing, querying, and interpreting the data from such archives. However, existing Vision-Language Models (VLMs) are predominantly trained on web-scraped, noisy image-text data, exhibiting limited exposure to the specialized domain of RS. This deficiency results in poor performance on RS-specific tasks, as commonly used datasets often lack detailed, scientifically accurate textual descriptions and instead emphasize solely on attributes like date and location. To bridge this critical gap, we introduce GAIA, a novel dataset designed for multi-scale, multi-sensor, and multi-modal RS image analysis. GAIA comprises of 205,150 meticulously curated RS image-text pairs, representing a diverse range of RS modalities associated to different spatial resolutions. Unlike existing vision-language datasets in RS, GAIA specifically focuses on capturing a diverse range of RS applications, providing unique information about environmental changes, natural disasters, and various other dynamic phenomena. The dataset provides a spatially and temporally balanced distribution, spanning across the globe, covering the last 25 years with a balanced temporal distribution of observations. GAIA's construction involved a two-stage process: (1) targeted web-scraping of images and accompanying text from reputable RS-related sources, and (2) generation of five high-quality, scientifically grounded synthetic captions for each image using carefully crafted prompts that leverage the advanced vision-language capabilities of GPT-4o. Our extensive experiments, including fine-tuning of CLIP and BLIP2 models, demonstrate that GAIA significantly improves performance on RS image classification, cross-modal retrieval and image captioning tasks.

Lars Möllenbrok, Behnood Rasti, Begüm Demir

Continual self-supervised learning (CSSL) methods have gained increasing attention in remote sensing (RS) due to their capability to learn new tasks sequentially from continuous streams of unlabeled data. Existing CSSL methods, while learning new tasks, focus on preventing catastrophic forgetting. To this end, most of them use regularization strategies to retain knowledge of previous tasks. This reduces the model's ability to adapt to the data of new tasks (i.e., learning plasticity), which can degrade performance. To address this problem, in this paper, we propose a novel CSSL method that aims to learn tasks sequentially, while achieving high learning plasticity. To this end, the proposed method uses a knowledge distillation strategy with an integrated decoupling mechanism. The decoupling is achieved by first dividing the feature dimensions into task-common and task-specific parts. Then, the task-common features are forced to be correlated to ensure memory stability while the task-specific features are forced to be de-correlated facilitating the learning of new features. Experimental results show the effectiveness of the proposed method compared to CaSSLe, which is a widely used CSSL framework, with improvements of up to 1.12% in average accuracy and 2.33% in intransigence in a task-incremental scenario, and 1.24% in average accuracy and 2.01% in intransigence in a class-incremental scenario.

Barış Büyüktaş, Gencer Sumbul, Begüm Demir

Federated learning (FL) enables the collaborative training of deep neural networks across decentralized data archives (i.e., clients) without sharing the local data of the clients. Most of the existing FL methods assume that the data distributed across all clients is associated with the same data modality. However, remote sensing (RS) images present in different clients can be associated with diverse data modalities. The joint use of the multi-modal RS data can significantly enhance classification performance. To effectively exploit decentralized and unshared multi-modal RS data, our paper introduces a novel multi-modal FL framework for RS image classification problems. The proposed framework comprises three modules: 1) multi-modal fusion (MF); 2) feature whitening (FW); and 3) mutual information maximization (MIM). The MF module employs iterative model averaging to facilitate learning without accessing multi-modal training data on clients. The FW module aims to address the limitations of training data heterogeneity by aligning data distributions across clients. The MIM module aims to model mutual information by maximizing the similarity between images from different modalities. For the experimental analyses, we focus our attention on multi-label classification and pixel-based classification tasks in RS. The results obtained using two benchmark archives show the effectiveness of the proposed framework when compared to state-of-the-art algorithms in the literature. The code of the proposed framework will be available at https://git.tu-berlin.de/rsim/multi-modal-FL.

Genc Hoxha, Olivér Angyal, Begüm Demir

The development of image time series retrieval (ITSR) methods is a growing research interest in remote sensing (RS). Given a user-defined image time series (i.e., the query time series), ITSR methods search and retrieve from large archives the image time series that have similar content to the query time series. Existing ITSR methods in RS are designed for unimodal retrieval problems, relying on an assumption that users always have access to a query image time series in the considered image modality. In operational scenarios, this assumption may not hold. To overcome this issue, as a first time in RS we introduce the task of cross-modal text-image time series retrieval (text-ITSR). In detail, we present a self-supervised cross-modal text-ITSR method that enables the retrieval of image time series using text sentences as queries, and vice versa. We focus our attention on text-ITSR in pairs of images (i.e., bitemporal images). Our text-ITSR method consists of two key components: 1) modality-specific encoders to model the semantic content of bitemporal images and text sentences with discriminative features; and 2) modality-specific projection heads to align textual and image representations in a shared embedding space. To effectively model the temporal information in the bitemporal images, we exploit two fusion strategies: i) global feature fusion (GFF) strategy that combines global image features through simple yet effective operators; and ii) transformer-based feature fusion (TFF) strategy that leverages transformers for fine-grained temporal integration. Extensive experiments conducted on two benchmark RS archives demonstrate the effectiveness of our method in accurately retrieving semantically relevant bitemporal images (or text sentences) to a query text sentence (or bitemporal image). The code of this work is publicly available at https://git.tu-berlin.de/rsim/cross-modal-text-tsir .

Binger Chen, Tacettin Emre Bök, Behnood Rasti et al.

Foundation Models (FMs) are increasingly used in remote sensing (RS) for tasks such as environmental monitoring, disaster assessment, and land-use mapping. These models include unimodal vision encoders trained on a single data modality and multimodal architectures trained on combinations of SAR, multispectral, hyperspectral, and image-text data. They support diverse RS tasks including semantic segmentation, image classification, change detection, and visual question answering. However, selecting an appropriate remote sensing foundation model (RSFM) remains difficult due to scattered documentation, heterogeneous formats, and varied deployment constraints. We introduce the RSFM Database (RS-FMD), a structured resource covering over 150 RSFMs spanning multiple data modalities, resolutions, and learning paradigms. Built on RS-FMD, we present REMSA, the first LLM-based agent for automated RSFM selection from natural language queries. REMSA interprets user requirements, resolves missing constraints, ranks candidate models using in-context learning, and provides transparent justifications. We also propose a benchmark of 75 expert-verified RS query scenarios, producing 900 configurations under an expert-centered evaluation protocol. REMSA outperforms several baselines, including naive agents, dense retrieval, and unstructured RAG-based LLMs. It operates entirely on publicly available metadata and does not access private or sensitive data.

Leonard Hackel, Tom Burgert, Begüm Demir

Self-supervised learning through masked autoencoders has attracted great attention for remote sensing (RS) foundation model (FM) development, enabling improved representation learning across diverse sensors and downstream tasks. However, existing RS FMs often either suffer from substantial computational complexity during both training and inference or exhibit limited representational capacity. These issues restrict their practical applicability in RS. To address this limitation, we propose an adaptation for enhancing the efficiency of RS FMs by integrating the Soft mixture-of-experts (MoE) mechanism into the FM. The integration of Soft MoEs into the FM allows modality-specific expert specialization alongside shared cross-sensor representation learning. To demonstrate the effectiveness of our adaptation, we apply it on the Cross-Sensor Masked Autoencoder (CSMAE) model, resulting in the Cross-Sensor Mixture-of-Experts (CSMoE) model. In addition, we introduce a thematic-climatic descriptor-driven sampling strategy for the construction of a representative and diverse training set to train our CSMoE model. Extensive experiments on scene classification, semantic segmentation, and content-based image retrieval demonstrate that our adaptation yields a reduction in computational requirements while maintaining or improving representational performance. Compared to state-of-the-art RS FMs, CSMoE achieves a superior trade-off between representational capacity, accuracy, and computational efficiency. On average, CSMoE achieves more than twice the computational efficiency of existing RS FMs, while maintaining competitive performance across all experiments. These results show the effectiveness of the proposed adaptation for creating computationally efficient RS FMs. The code for the model, the training set creation, and the model weights will be available at https://git.tu-berlin.de/rsim/csmoe.

Maximilian Kromer, Panagiotis Agrafiotis, Begüm Demir

Accurate image-based bathymetric mapping in shallow waters remains challenging due to the complex optical distortions such as wave induced patterns, scattering and sunglint, introduced by the dynamic water surface, the water column properties, and solar illumination. In this work, we introduce Sea-Undistort, a comprehensive synthetic dataset of 1200 paired 512x512 through-water scenes rendered in Blender. Each pair comprises a distortion-free and a distorted view, featuring realistic water effects such as sun glint, waves, and scattering over diverse seabeds. Accompanied by per-image metadata such as camera parameters, sun position, and average depth, Sea-Undistort enables supervised training that is otherwise infeasible in real environments. We use Sea-Undistort to benchmark two state-of-the-art image restoration methods alongside an enhanced lightweight diffusion-based framework with an early-fusion sun-glint mask. When applied to real aerial data, the enhanced diffusion model delivers more complete Digital Surface Models (DSMs) of the seabed, especially in deeper areas, reduces bathymetric errors, suppresses glint and scattering, and crisply restores fine seabed details. Dataset, weights, and code are publicly available at https://www.magicbathy.eu/Sea-Undistort.html.

Barış Büyüktaş, Jonas Klotz, Begüm Demir

Federated learning (FL) enables the collaborative training of deep neural networks across decentralized data archives (i.e., clients), where each client stores data locally and only shares model updates with a central server. This makes FL a suitable learning paradigm for remote sensing (RS) image classification tasks, where data centralization may be restricted due to legal and privacy constraints. However, a key challenge in applying FL to RS tasks is the communication overhead caused by the frequent exchange of large model updates between clients and the central server. To address this issue, in this paper we propose a novel strategy (denoted as FedX) that uses explanation-guided pruning to reduce communication overhead by minimizing the size of the transmitted models without compromising performance. FedX leverages backpropagation-based explanation methods to estimate the task-specific importance of model components and prunes the least relevant ones at the central server. The resulting sparse global model is then sent to clients, substantially reducing communication overhead. We evaluate FedX on multi-label scene classification using the BigEarthNet-S2 dataset and single-label scene classification using the EuroSAT dataset. Experimental results show the success of FedX in significantly reducing the number of shared model parameters while enhancing the generalization capability of the global model, compared to both unpruned model and state-of-the-art pruning methods. The code of FedX will be available at https://git.tu-berlin.de/rsim/FedX.

Jonas Klotz, Tom Burgert, Begüm Demir

The development of explainable artificial intelligence (xAI) methods for scene classification problems has attracted great attention in remote sensing (RS). Most xAI methods and the related evaluation metrics in RS are initially developed for natural images considered in computer vision (CV), and their direct usage in RS may not be suitable. To address this issue, in this paper, we investigate the effectiveness of explanation methods and metrics in the context of RS image scene classification. In detail, we methodologically and experimentally analyze ten explanation metrics spanning five categories (faithfulness, robustness, localization, complexity, randomization), applied to five established feature attribution methods (Occlusion, LIME, GradCAM, LRP, and DeepLIFT) across three RS datasets. Our methodological analysis identifies key limitations in both explanation methods and metrics. The performance of perturbation-based methods, such as Occlusion and LIME, heavily depends on perturbation baselines and spatial characteristics of RS scenes. Gradient-based approaches like GradCAM struggle when multiple labels are present in the same image, while some relevance propagation methods (LRP) can distribute relevance disproportionately relative to the spatial extent of classes. Analogously, we find limitations in evaluation metrics. Faithfulness metrics share the same problems as perturbation-based methods. Localization metrics and complexity metrics are unreliable for classes with a large spatial extent. In contrast, robustness metrics and randomization metrics consistently exhibit greater stability. Our experimental results support these methodological findings. Based on our analysis, we provide guidelines for selecting explanation methods, metrics, and hyperparameters in the context of RS image scene classification.

Yan Shu, Bin Ren, Zhitong Xiong et al.

Earth Observation (EO) data analysis is vital for monitoring environmental and human dynamics. Recent Multimodal Large Language Models (MLLMs) show potential in EO understanding but remain restricted to single-sensor inputs, overlooking the complementarity across heterogeneous modalities. We propose EarthMind, a unified vision-language framework that handles both single- and cross-sensor inputs via an innovative hierarchical cross-modal attention (ie, HCA) design. Specifically, HCA hierarchically captures visual relationships across sensors and aligns them with language queries, enabling adaptive fusion of optical and Synthetic Aperture Radar (SAR) features. To support cross-sensor learning, we curate FusionEO, a 30K-pair dataset with diverse annotations, and establish EarthMind-Bench, a 2,841-pair benchmark with expert annotations for perception and reasoning tasks. Extensive experiments show that EarthMind achieves state-of-the-art results on EarthMind-Bench and surpasses existing MLLMs on multiple EO benchmarks.

Jonas Klotz, Barış Büyüktaş, Begüm Demir

Federated learning (FL) is a decentralized machine learning paradigm in which multiple clients collaboratively train a global model by exchanging only model updates with the central server without sharing the local data of the clients. Due to the large volume of model updates required to be transmitted between clients and the central server, most FL systems are associated with high transfer costs (i.e., communication overhead). This issue is more critical for operational applications in remote sensing (RS), especially when large-scale RS data is processed and analyzed through FL systems with restricted communication bandwidth. To address this issue, we introduce an explanation-guided pruning strategy for communication-efficient FL in the context of RS image classification. Our pruning strategy is defined based on the layer-wise relevance propagation (LRP) driven explanations to: 1) efficiently and effectively identify the most relevant and informative model parameters (to be exchanged between clients and the central server); and 2) eliminate the non-informative ones to minimize the volume of model updates. The experimental results on the BigEarthNet-S2 dataset demonstrate that our strategy effectively reduces the number of shared model updates, while increasing the generalization ability of the global model. The code of this work is publicly available at https://git.tu-berlin.de/rsim/FL-LRP.

Theresa Follath, David Mickisch, Jan Hemmerling et al.

Using images acquired by different satellite sensors has shown to improve classification performance in the framework of crop mapping from satellite image time series (SITS). Existing state-of-the-art architectures use self-attention mechanisms to process the temporal dimension and convolutions for the spatial dimension of SITS. Motivated by the success of purely attention-based architectures in crop mapping from single-modal SITS, we introduce several multi-modal multi-temporal transformer-based architectures. Specifically, we investigate the effectiveness of Early Fusion, Cross Attention Fusion and Synchronized Class Token Fusion within the Temporo-Spatial Vision Transformer (TSViT). Experimental results demonstrate significant improvements over state-of-the-art architectures with both convolutional and self-attention components.

Genc Hoxha, Gencer Sumbul, Julia Henkel et al.

Deep metric learning (DML) has shown to be effective for content-based image retrieval (CBIR) in remote sensing (RS). Most of DML methods for CBIR rely on a high number of annotated images to accurately learn model parameters of deep neural networks (DNNs). However, gathering such data is time-consuming and costly. To address this, we propose an annotation cost-efficient active learning (ANNEAL) method tailored to DML-driven CBIR in RS. ANNEAL aims to create a small but informative training set made up of similar and dissimilar image pairs to be utilized for accurately learning a metric space. The informativeness of image pairs is evaluated by combining uncertainty and diversity criteria. To assess the uncertainty of image pairs, we introduce two algorithms: 1) metric-guided uncertainty estimation (MGUE); and 2) binary classifier guided uncertainty estimation (BCGUE). MGUE algorithm automatically estimates a threshold value that acts as a boundary between similar and dissimilar image pairs based on the distances in the metric space. The closer the similarity between image pairs is to the estimated threshold value the higher their uncertainty. BCGUE algorithm estimates the uncertainty of the image pairs based on the confidence of the classifier in assigning correct similarity labels. The diversity criterion is assessed through a clustering-based strategy. ANNEAL combines either MGUE or BCGUE algorithm with the clustering-based strategy to select the most informative image pairs, which are then labelled by expert annotators as similar or dissimilar. This way of annotating images significantly reduces the annotation cost compared to annotating images with land-use land-cover class labels. Experimental results on two RS benchmark datasets demonstrate the effectiveness of our method. The code of this work is publicly available at \url{https://git.tu-berlin.de/rsim/anneal_tgrs}.

Martin Hermann Paul Fuchs, Akshara Preethy Byju, Alisa Walda et al.

The development of deep learning-based models for the compression of hyperspectral images (HSIs) has recently attracted great attention in remote sensing due to the sharp growing of hyperspectral data archives. Most of the existing models achieve either spectral or spatial compression, and do not jointly consider the spatio-spectral redundancies present in HSIs. To address this problem, in this paper we focus our attention on the High Fidelity Compression (HiFiC) model (which is proven to be highly effective for spatial compression problems) and adapt it to perform spatio-spectral compression of HSIs. In detail, we introduce two new models: i) HiFiC using Squeeze and Excitation (SE) blocks (denoted as HiFiC$_{SE}$); and ii) HiFiC with 3D convolutions (denoted as HiFiC$_{3D}$) in the framework of compression of HSIs. We analyze the effectiveness of HiFiC$_{SE}$ and HiFiC$_{3D}$ in compressing the spatio-spectral redundancies with channel attention and inter-dependency analysis. Experimental results show the efficacy of the proposed models in performing spatio-spectral compression, while reconstructing images at reduced bitrates with higher reconstruction quality. The code of the proposed models is publicly available at https://git.tu-berlin.de/rsim/HSI-SSC .

Devis Tuia, Konrad Schindler, Begüm Demir et al.

Earth observation (EO) is a prime instrument for monitoring land and ocean processes, studying the dynamics at work, and taking the pulse of our planet. This article gives a bird's eye view of the essential scientific tools and approaches informing and supporting the transition from raw EO data to usable EO-based information. The promises, as well as the current challenges of these developments, are highlighted under dedicated sections. Specifically, we cover the impact of (i) Computer vision; (ii) Machine learning; (iii) Advanced processing and computing; (iv) Knowledge-based AI; (v) Explainable AI and causal inference; (vi) Physics-aware models; (vii) User-centric approaches; and (viii) the much-needed discussion of ethical and societal issues related to the massive use of ML technologies in EO.

Georgii Mikriukov, Mahdyar Ravanbakhsh, Begüm Demir

The development of accurate and scalable cross-modal image-text retrieval methods, where queries from one modality (e.g., text) can be matched to archive entries from another (e.g., remote sensing image) has attracted great attention in remote sensing (RS). Most of the existing methods assume that a reliable multi-modal training set with accurately matched text-image pairs is existing. However, this assumption may not always hold since the multi-modal training sets may include noisy pairs (i.e., textual descriptions/captions associated to training images can be noisy), distorting the learning process of the retrieval methods. To address this problem, we propose a novel unsupervised cross-modal hashing method robust to the noisy image-text correspondences (CHNR). CHNR consists of three modules: 1) feature extraction module, which extracts feature representations of image-text pairs; 2) noise detection module, which detects potential noisy correspondences; and 3) hashing module that generates cross-modal binary hash codes. The proposed CHNR includes two training phases: i) meta-learning phase that uses a small portion of clean (i.e., reliable) data to train the noise detection module in an adversarial fashion; and ii) the main training phase for which the trained noise detection module is used to identify noisy correspondences while the hashing module is trained on the noisy multi-modal training set. Experimental results show that the proposed CHNR outperforms state-of-the-art methods. Our code is publicly available at https://git.tu-berlin.de/rsim/chnr