Bertrand Le Saux

Su Yeon Chang, Michele Grossi, Bertrand Le Saux et al.

Quantum Neural Networks (QNNs) are suggested as one of the quantum algorithms which can be efficiently simulated with a low depth on near-term quantum hardware in the presence of noises. However, their performance highly relies on choosing the most suitable architecture of Variational Quantum Algorithms (VQAs), and the problem-agnostic models often suffer issues regarding trainability and generalization power. As a solution, the most recent works explore Geometric Quantum Machine Learning (GQML) using QNNs equivariant with respect to the underlying symmetry of the dataset. GQML adds an inductive bias to the model by incorporating the prior knowledge on the given dataset and leads to enhancing the optimization performance while constraining the search space. This work proposes equivariant Quantum Convolutional Neural Networks (EquivQCNNs) for image classification under planar $p4m$ symmetry, including reflectional and $90^\circ$ rotational symmetry. We present the results tested in different use cases, such as phase detection of the 2D Ising model and classification of the extended MNIST dataset, and compare them with those obtained with the non-equivariant model, proving that the equivariance fosters better generalization of the model.

Amer Delilbasic, Bertrand Le Saux, Morris Riedel et al.

In recent years, the development of quantum annealers has enabled experimental demonstrations and has increased research interest in applications of quantum annealing, such as in quantum machine learning and in particular for the popular quantum SVM. Several versions of the quantum SVM have been proposed, and quantum annealing has been shown to be effective in them. Extensions to multiclass problems have also been made, which consist of an ensemble of multiple binary classifiers. This work proposes a novel quantum SVM formulation for direct multiclass classification based on quantum annealing, called Quantum Multiclass SVM (QMSVM). The multiclass classification problem is formulated as a single Quadratic Unconstrained Binary Optimization (QUBO) problem solved with quantum annealing. The main objective of this work is to evaluate the feasibility, accuracy, and time performance of this approach. Experiments have been performed on the D-Wave Advantage quantum annealer for a classification problem on remote sensing data. The results indicate that, despite the memory demands of the quantum annealer, QMSVM can achieve accuracy that is comparable to standard SVM methods and, more importantly, it scales much more efficiently with the number of training examples, resulting in nearly constant time. This work shows an approach for bringing together classical and quantum computation, solving practical problems in remote sensing with current hardware.

Christel Chappuis, Eliot Walt, Vincent Mendez et al.

Remote sensing visual question answering (RSVQA) opens new opportunities for the use of overhead imagery by the general public, by enabling human-machine interaction with natural language. Building on the recent advances in natural language processing and computer vision, the goal of RSVQA is to answer a question formulated in natural language about a remote sensing image. Language understanding is essential to the success of the task, but has not yet been thoroughly examined in RSVQA. In particular, the problem of language biases is often overlooked in the remote sensing community, which can impact model robustness and lead to wrong conclusions about the performances of the model. Thus, the present work aims at highlighting the problem of language biases in RSVQA with a threefold analysis strategy: visual blind models, adversarial testing and dataset analysis. This analysis focuses both on model and data. Moreover, we motivate the use of more informative and complementary evaluation metrics sensitive to the issue. The gravity of language biases in RSVQA is then exposed for all of these methods with the training of models discarding the image data and the manipulation of the visual input during inference. Finally, a detailed analysis of question-answer distribution demonstrates the root of the problem in the data itself. Thanks to this analytical study, we observed that biases in remote sensing are more severe than in standard VQA, likely due to the specifics of existing remote sensing datasets for the task, e.g. geographical similarities and sparsity, as well as a simpler vocabulary and question generation strategies. While new, improved and less-biased datasets appear as a necessity for the development of the promising field of RSVQA, we demonstrate that more informed, relative evaluation metrics remain much needed to transparently communicate results of future RSVQA methods.

Artur Miroszewski, Jakub Mielczarek, Grzegorz Czelusta et al.

Support vector machines (SVMs) are a well-established classifier effectively deployed in an array of classification tasks. In this work, we consider extending classical SVMs with quantum kernels and applying them to satellite data analysis. The design and implementation of SVMs with quantum kernels (hybrid SVMs) are presented. Here, the pixels are mapped to the Hilbert space using a family of parameterized quantum feature maps (related to quantum kernels). The parameters are optimized to maximize the kernel target alignment. The quantum kernels have been selected such that they enabled analysis of numerous relevant properties while being able to simulate them with classical computers on a real-life large-scale dataset. Specifically, we approach the problem of cloud detection in the multispectral satellite imagery, which is one of the pivotal steps in both on-the-ground and on-board satellite image analysis processing chains. The experiments performed over the benchmark Landsat-8 multispectral dataset revealed that the simulated hybrid SVM successfully classifies satellite images with accuracy comparable to the classical SVM with the RBF kernel for large datasets. Interestingly, for large datasets, the high accuracy was also observed for the simple quantum kernels, lacking quantum entanglement.

Michał Siemaszko, Adam Buraczewski, Bertrand Le Saux et al.

Time series prediction is essential for human activities in diverse areas. A common approach to this task is to harness Recurrent Neural Networks (RNNs). However, while their predictions are quite accurate, their learning process is complex and, thus, time and energy consuming. Here, we propose to extend the concept of RRNs by including continuous-variable quantum resources in it, and to use a quantum-enhanced RNN to overcome these obstacles. The design of the Continuous-Variable Quantum RNN (CV-QRNN) is rooted in the continuous-variable quantum computing paradigm. By performing extensive numerical simulations, we demonstrate that the quantum network is capable of learning-time dependence of several types of temporal data, and that it converges to the optimal weights in fewer epochs than a classical network. Furthermore, for a small number of trainable parameters, it can achieve lower losses than its classical counterpart. CV-QRNN can be implemented using commercially available quantum-photonic hardware.

Lukasz Tulczyjew, Michal Kawulok, Nicolas Longépé et al.

Hyperspectral unmixing remains one of the most challenging tasks in the analysis of such data. Deep learning has been blooming in the field and proved to outperform other classic unmixing techniques, and can be effectively deployed onboard Earth observation satellites equipped with hyperspectral imagers. In this letter, we follow this research pathway and propose a multi-branch convolutional neural network that benefits from fusing spectral, spatial, and spectral-spatial features in the unmixing process. The results of our experiments, backed up with the ablation study, revealed that our techniques outperform others from the literature and lead to higher-quality fractional abundance estimation. Also, we investigated the influence of reducing the training sets on the capabilities of all algorithms and their robustness against noise, as capturing large and representative ground-truth sets is time-consuming and costly in practice, especially in emerging Earth observation scenarios.

Lukasz Tulczyjew, Michal Kawulok, Nicolas Longépé et al.

Maintaining farm sustainability through optimizing the agricultural management practices helps build more planet-friendly environment. The emerging satellite missions can acquire multi- and hyperspectral imagery which captures more detailed spectral information concerning the scanned area, hence allows us to benefit from subtle spectral features during the analysis process in agricultural applications. We introduce an approach for extracting 2.5 m cultivated land maps from 10 m Sentinel-2 multispectral image series which benefits from a compact graph convolutional neural network. The experiments indicate that our models not only outperform classical and deep machine learning techniques through delivering higher-quality segmentation maps, but also dramatically reduce the memory footprint when compared to U-Nets (almost 8k trainable parameters of our models, with up to 31M parameters of U-Nets). Such memory frugality is pivotal in the missions which allow us to uplink a model to the AI-powered satellite once it is in orbit, as sending large nets is impossible due to the time constraints.

Michael L. Marszalek, Bertrand Le Saux, Pierre-Philippe Mathieu et al.

Machine learning, satellites or local sensors are key factors for a sustainable and resource-saving optimisation of agriculture and proved its values for the management of agricultural land. Up to now, the main focus was on the enlargement of data which were evaluated by means of supervised learning methods. Nevertheless, the need for labels is also a limiting and time-consuming factor, while in contrast, ongoing technological development is already providing an ever-increasing amount of unlabeled data. Self-supervised learning (SSL) could overcome this limitation and incorporate existing unlabeled data. Therefore, a crop type data set was utilized to conduct experiments with SSL and compare it to supervised methods. A unique feature of our data set from 2016 to 2018 was a divergent climatological condition in 2018 that reduced yields and affected the spectral fingerprint of the plants. Our experiments focused on predicting 2018 using SLL without or a few labels to clarify whether new labels should be collected for an unknown year. Despite these challenging conditions, the results showed that SSL contributed to higher accuracies. We believe that the results will encourage further improvements in the field of precision farming, why the SSL framework and data will be published (Marszalek, 2021).

Artur Miroszewski, Jakub Mielczarek, Filip Szczepanek et al.

Support vector machines (SVMs) are a well-established classifier effectively deployed in an array of pattern recognition and classification tasks. In this work, we consider extending classic SVMs with quantum kernels and applying them to satellite data analysis. The design and implementation of SVMs with quantum kernels (hybrid SVMs) is presented. It consists of the Quantum Kernel Estimation (QKE) procedure combined with a classic SVM training routine. The pixel data are mapped to the Hilbert space using ZZ-feature maps acting on the parameterized ansatz state. The parameters are optimized to maximize the kernel target alignment. We approach the problem of cloud detection in satellite image data, which is one of the pivotal steps in both on-the-ground and on-board satellite image analysis processing chains. The experiments performed over the benchmark Landsat-8 multispectral dataset revealed that the simulated hybrid SVM successfully classifies satellite images with accuracy on par with classic SVMs.

Andrea Ceschini, Francesco Mauro, Francesca De Falco et al.

Quantum Graph Neural Networks (QGNNs) represent a novel fusion of quantum computing and Graph Neural Networks (GNNs), aimed at overcoming the computational and scalability challenges inherent in classical GNNs that are powerful tools for analyzing data with complex relational structures but suffer from limitations such as high computational complexity and over-smoothing in large-scale applications. Quantum computing, leveraging principles like superposition and entanglement, offers a pathway to enhanced computational capabilities. This paper critically reviews the state-of-the-art in QGNNs, exploring various architectures. We discuss their applications across diverse fields such as high-energy physics, molecular chemistry, finance and earth sciences, highlighting the potential for quantum advantage. Additionally, we address the significant challenges faced by QGNNs, including noise, decoherence, and scalability issues, proposing potential strategies to mitigate these problems. This comprehensive review aims to provide a foundational understanding of QGNNs, fostering further research and development in this promising interdisciplinary field.

Bartosz Grabowski, Maciej Ziaja, Michal Kawulok et al.

Cloud detection is a pivotal satellite image pre-processing step that can be performed both on the ground and on board a satellite to tag useful images. In the latter case, it can reduce the amount of data to downlink by pruning the cloudy areas, or to make a satellite more autonomous through data-driven acquisition re-scheduling. We approach this task with nnU-Nets, a self-reconfigurable framework able to perform meta-learning of a segmentation network over various datasets. Unfortunately, such models are commonly memory-inefficient due to their (very) large architectures. To benefit from them in on-board processing, we compress nnU-Nets with knowledge distillation into much smaller and compact U-Nets. Our experiments, performed over Sentinel-2 and Landsat-8 images revealed that nnU-Nets deliver state-of-the-art performance without any manual design. Our approach was ranked within the top 7% best solutions (across 847 teams) in the On Cloud N: Cloud Cover Detection Challenge, where we reached the Jaccard index of 0.882 over more than 10k unseen Sentinel-2 images (the winners obtained 0.897, the baseline U-Net with the ResNet-34 backbone: 0.817, and the classic Sentinel-2 image thresholding: 0.652). Finally, we showed that knowledge distillation enables to elaborate dramatically smaller (almost 280x) U-Nets when compared to nnU-Nets while still maintaining their segmentation capabilities.

Bartosz Grabowski, Maciej Ziaja, Michal Kawulok et al.

Cloud detection is a pivotal satellite image pre-processing step that can be performed both on the ground and on board a satellite to tag useful images. In the latter case, it can help to reduce the amount of data to downlink by pruning the cloudy areas, or to make a satellite more autonomous through data-driven acquisition re-scheduling of the cloudy areas. We approach this important task with nnU-Nets, a self-reconfigurable framework able to perform meta-learning of a segmentation network over various datasets. Our experiments, performed over Sentinel-2 and Landsat-8 multispectral images revealed that nnU-Nets deliver state-of-the-art cloud segmentation performance without any manual design. Our approach was ranked within the top 7% best solutions (across 847 participating teams) in the On Cloud N: Cloud Cover Detection Challenge, where we reached the Jaccard index of 0.882 over more than 10k unseen Sentinel-2 image patches (the winners obtained 0.897, whereas the baseline U-Net with the ResNet-34 backbone used as an encoder: 0.817, and the classic Sentinel-2 image thresholding: 0.652).

Artur Miroszewski, Jakub Mielczarek, Filip Szczepanek et al.

The optimization of Kernel-Target Alignment (TA) has been recently proposed as a way to reduce the number of hardware resources in quantum classifiers. It allows to exchange highly expressive and costly circuits to moderate size, task oriented ones. In this work we propose a simple toy model to study the optimization landscape of the Kernel-Target Alignment. We find that for underparameterized circuits the optimization landscape possess either many local extrema or becomes flat with narrow global extremum. We find the dependence of the width of the global extremum peak on the amount of data introduced to the model. The experimental study was performed using multispectral satellite data, and we targeted the cloud detection task, being one of the most fundamental and important image analysis tasks in remote sensing.

Bálint Máté, Bertrand Le Saux, Maxwell Henderson

This paper explores the advantages of optimizing quantum circuits on $N$ wires as operators in the unitary group $U(2^N)$. We run gradient-based optimization in the Lie algebra $\mathfrak u(2^N)$ and use the exponential map to parametrize unitary matrices. We argue that $U(2^N)$ is not only more general than the search space induced by an ansatz, but in ways easier to work with on classical computers. The resulting approach is quick, ansatz-free and provides an upper bound on performance over all ansätze on $N$ wires.

Kai Jeggle, Mikolaj Czerkawski, Federico Serva et al.

Clouds containing ice particles play a crucial role in the climate system. Yet they remain a source of great uncertainty in climate models and future climate projections. In this work, we create a new observational constraint of regime-dependent ice microphysical properties at the spatio-temporal coverage of geostationary satellite instruments and the quality of active satellite retrievals. We achieve this by training a convolutional neural network on three years of SEVIRI and DARDAR data sets. This work will enable novel research to improve ice cloud process understanding and hence, reduce uncertainties in a changing climate and help assess geoengineering methods for cirrus clouds.

Artur Miroszewski, Jakub Nalepa, Bertrand Le Saux et al.

The industry of quantum technologies is rapidly expanding, offering promising opportunities for various scientific domains. Among these emerging technologies, Quantum Machine Learning (QML) has attracted considerable attention due to its potential to revolutionize data processing and analysis. In this paper, we investigate the application of QML in the field of remote sensing. It is believed that QML can provide valuable insights for analysis of data from space. We delve into the common beliefs surrounding the quantum advantage in QML for remote sensing and highlight the open challenges that need to be addressed. To shed light on the challenges, we conduct a study focused on the problem of kernel value concentration, a phenomenon that adversely affects the runtime of quantum computers. Our findings indicate that while this issue negatively impacts quantum computer performance, it does not entirely negate the potential quantum advantage in QML for remote sensing.

Fulvio Sanguigni, Mikolaj Czerkawski, Lorenzo Papa et al.

The advancements in the state of the art of generative Artificial Intelligence (AI) brought by diffusion models can be highly beneficial in novel contexts involving Earth observation data. After introducing this new family of generative models, this work proposes and analyses three use cases which demonstrate the potential of diffusion-based approaches for satellite image data. Namely, we tackle cloud removal and inpainting, dataset generation for change-detection tasks, and urban replanning.

Zhitong Xiong, Yi Wang, Fahong Zhang et al.

Earth observation (EO) in open-world settings presents a unique challenge: different applications rely on diverse sensor modalities, each with varying ground sampling distances, spectral ranges, and numbers of spectral bands. However, existing EO foundation models are typically tailored to specific sensor types, making them inflexible when generalizing across the heterogeneous landscape of EO data. To address this, we propose the Dynamic One-For-All (DOFA) model, a unified, multimodal foundation framework designed for diverse vision tasks in EO. Inspired by neural plasticity, DOFA utilizes a wavelength-conditioned dynamic hypernetwork to process inputs from five distinct satellite sensors flexibly. By continually pretraining on five EO modalities, DOFA achieves state-of-the-art performance across multiple downstream tasks and generalizes well to unseen modalities. Enhanced with hybrid continual pretraining, DOFA+ requires significantly fewer computational resources while outperforming counterparts trained with extensive GPU budgets. Experiments on diverse datasets highlight DOFA's potential as a foundation for general-purpose vision models in the sensor-diverse EO domain. The code and pre-trained weights are publicly available at https://github.com/zhu-xlab/DOFA.

Artur Miroszewski, Marco Fellous Asiani, Jakub Mielczarek et al.

Quantum Machine Learning (QML) has gathered significant attention through approaches like Quantum Kernel Machines. While these methods hold considerable promise, their quantum nature presents inherent challenges. One major challenge is the limited resolution of estimated kernel values caused by the finite number of circuit runs performed on a quantum device. In this study, we propose a comprehensive system of rules and heuristics for estimating the required number of circuit runs in quantum kernel methods. We introduce two critical effects that necessitate an increased measurement precision through additional circuit runs: the spread effect and the concentration effect. The effects are analyzed in the context of fidelity and projected quantum kernels. To address these phenomena, we develop an approach for estimating desired precision of kernel values, which, in turn, is translated into the number of circuit runs. Our methodology is validated through extensive numerical simulations, focusing on the problem of exponential value concentration. We stress that quantum kernel methods should not only be considered from the machine learning performance perspective, but also from the context of the resource consumption. The results provide insights into the possible benefits of quantum kernel methods, offering a guidance for their application in quantum machine learning tasks.

Daniel Paluba, Bertrand Le Saux, Přemysl Stych

Current optical vegetation indices (VIs) for monitoring forest ecosystems are well established and widely used in various applications, but can be limited by atmospheric effects such as clouds. In contrast, synthetic aperture radar (SAR) data can offer insightful and systematic forest monitoring with complete time series (TS) due to signal penetration through clouds and day and night image acquisitions. This study aims to address the limitations of optical satellite data by using SAR data as an alternative for estimating optical VIs for forests through machine learning (ML). While this approach is less direct and likely only feasible through the power of ML, it raises the scientific question of whether enough relevant information is contained in the SAR signal to accurately estimate VIs. This work covers the estimation of TS of four VIs (LAI, FAPAR, EVI and NDVI) using multitemporal Sentinel-1 SAR and ancillary data. The study focused on both healthy and disturbed temperate forest areas in Czechia for the year 2021, while ground truth labels generated from Sentinel-2 multispectral data. This was enabled by creating a paired multi-modal TS dataset in Google Earth Engine (GEE), including temporally and spatially aligned Sentinel-1, Sentinel-2, DEM, weather and land cover datasets. The inclusion of DEM-derived auxiliary features and additional meteorological information, further improved the results. In the comparison of ML models, the traditional ML algorithms, RFR and XGBoost slightly outperformed the AutoML approach, auto-sklearn, for all VIs, achieving high accuracies ($R^2$ between 70-86%) and low errors (0.055-0.29 of MAE). In general, up to 240 measurements per year and a spatial resolution of 20 m can be achieved using estimated SAR-based VIs with high accuracy. A great advantage of the SAR-based VI is the ability to detect abrupt forest changes with sub-weekly temporal accuracy.

Gaston Lenczner, Adrien Chan-Hon-Tong, Nicola Luminari et al.

Transfer learning is a powerful way to adapt existing deep learning models to new emerging use-cases in remote sensing. Starting from a neural network already trained for semantic segmentation, we propose to modify its label space to swiftly adapt it to new classes under weak supervision. To alleviate the background shift and the catastrophic forgetting problems inherent to this form of continual learning, we compare different regularization terms and leverage a pseudo-label strategy. We experimentally show the relevance of our approach on three public remote sensing datasets. Code is open-source and released in this repository: https://github.com/alteia-ai/ICSS}{https://github.com/alteia-ai/ICSS.

Daniela A. Zaidenberg, Alessandro Sebastianelli, Dario Spiller et al.

This concept paper aims to provide a brief outline of quantum computers, explore existing methods of quantum image classification techniques, so focusing on remote sensing applications, and discuss the bottlenecks of performing these algorithms on currently available open source platforms. Initial results demonstrate feasibility. Next steps include expanding the size of the quantum hidden layer and increasing the variety of output image options.

Gaston Lenczner, Bertrand Le Saux, Nicola Luminari et al.

This paper presents an interactive approach for multi-class segmentation of aerial images. Precisely, it is based on a deep neural network which exploits both RGB images and annotations. Starting from an initial output based on the image only, our network then interactively refines this segmentation map using a concatenation of the image and user annotations. Importantly, user annotations modify the inputs of the network - not its weights - enabling a fast and smooth process. Through experiments on two public aerial datasets, we show that user annotations are extremely rewarding: each click corrects roughly 5000 pixels. We analyze the impact of different aspects of our framework such as the representation of the annotations, the volume of training data or the network architecture. Code is available at https://github.com/delair-ai/DISIR.

Marcela Carvalho, Bertrand Le Saux, Pauline Trouvé-Peloux et al.

Aerial or satellite imagery is a great source for land surface analysis, which might yield land use maps or elevation models. In this investigation, we present a neural network framework for learning semantics and local height together. We show how this joint multi-task learning benefits to each task on the large dataset of the 2018 Data Fusion Contest. Moreover, our framework also yields an uncertainty map which allows assessing the prediction of the model. Code is available at https://github.com/marcelampc/mtl_aerial_images .

Marcela Carvalho, Bertrand Le Saux, Pauline Trouvé-Peloux et al.

Depth estimation is of critical interest for scene understanding and accurate 3D reconstruction. Most recent approaches in depth estimation with deep learning exploit geometrical structures of standard sharp images to predict corresponding depth maps. However, cameras can also produce images with defocus blur depending on the depth of the objects and camera settings. Hence, these features may represent an important hint for learning to predict depth. In this paper, we propose a full system for single-image depth prediction in the wild using depth-from-defocus and neural networks. We carry out thorough experiments to test deep convolutional networks on real and simulated defocused images using a realistic model of blur variation with respect to depth. We also investigate the influence of blur on depth prediction observing model uncertainty with a Bayesian neural network approach. From these studies, we show that out-of-focus blur greatly improves the depth-prediction network performances. Furthermore, we transfer the ability learned on a synthetic, indoor dataset to real, indoor and outdoor images. For this purpose, we present a new dataset containing real all-focus and defocused images from a Digital Single-Lens Reflex (DSLR) camera, paired with ground truth depth maps obtained with an active 3D sensor for indoor scenes. The proposed approach is successfully validated on both this new dataset and standard ones as NYUv2 or Depth-in-the-Wild. Code and new datasets are available at https://github.com/marcelampc/d3net_depth_estimation

Casper Fibaek, Luke Camilleri, Andreas Luyts et al.

Massive amounts of unlabelled data are captured by Earth Observation (EO) satellites, with the Sentinel-2 constellation generating 1.6 TB of data daily. This makes Remote Sensing a data-rich domain well suited to Machine Learning (ML) solutions. However, a bottleneck in applying ML models to EO is the lack of annotated data as annotation is a labour-intensive and costly process. As a result, research in this domain has focused on Self-Supervised Learning and Foundation Model approaches. This paper addresses the need to evaluate different Foundation Models on a fair and uniform benchmark by introducing the PhilEO Bench, a novel evaluation framework for EO Foundation Models. The framework comprises of a testbed and a novel 400 GB Sentinel-2 dataset containing labels for three downstream tasks, building density estimation, road segmentation, and land cover classification. We present experiments using our framework evaluating different Foundation Models, including Prithvi and SatMAE, at multiple n-shots and convergence rates.

Francesco Pro, Nikolaos Dionelis, Luca Maiano et al.

Nowadays the accurate geo-localization of ground-view images has an important role across domains as diverse as journalism, forensics analysis, transports, and Earth Observation. This work addresses the problem of matching a query ground-view image with the corresponding satellite image without GPS data. This is done by comparing the features from a ground-view image and a satellite one, innovatively leveraging the corresponding latter's segmentation mask through a three-stream Siamese-like network. The proposed method, Semantic Align Net (SAN), focuses on limited Field-of-View (FoV) and ground panorama images (images with a FoV of 360°). The novelty lies in the fusion of satellite images in combination with their semantic segmentation masks, aimed at ensuring that the model can extract useful features and focus on the significant parts of the images. This work shows how SAN through semantic analysis of images improves the performance on the unlabelled CVUSA dataset for all the tested FoVs.

Nikolaos Dionelis, Francesco Pro, Luca Maiano et al.

Data from satellites or aerial vehicles are most of the times unlabelled. Annotating such data accurately is difficult, requires expertise, and is costly in terms of time. Even if Earth Observation (EO) data were correctly labelled, labels might change over time. Learning from unlabelled data within a semi-supervised learning framework for segmentation of aerial images is challenging. In this paper, we develop a new model for semantic segmentation of unlabelled images, the Non-annotated Earth Observation Semantic Segmentation (NEOS) model. NEOS performs domain adaptation as the target domain does not have ground truth semantic segmentation masks. The distribution inconsistencies between the target and source domains are due to differences in acquisition scenes, environment conditions, sensors, and times. Our model aligns the learned representations of the different domains to make them coincide. The evaluation results show that NEOS is successful and outperforms other models for semantic segmentation of unlabelled data.

Nikolaos Dionelis, Riccardo Musto, Jente Bosmans et al.

Today, Earth Observation (EO) satellites generate massive volumes of data. To fully exploit this, it is essential to pretrain EO Foundation Models (FMs) on large unlabeled datasets, enabling efficient fine-tuning for downstream tasks with minimal labeled data. In this paper, we study scaling-up FMs: we train our models on the pretraining dataset MajorTOM 23TB which includes all regions, and the performance on average is competitive versus models pretrained on more specialized datasets which are substantially smaller and include only land. The additional data of oceans and ice do not decrease the performance on land-focused downstream tasks. These results indicate that large FMs trained on global datasets for a wider variety of downstream tasks can be useful for downstream applications that only require a subset of the information included in their training. The second contribution is the exploration of U-Net Convolutional Neural Network (CNN), Vision Transformers (ViT), and Mamba State-Space Models (SSM) as FMs. U-Net captures local correlations amongst pixels, while ViT and Mamba capture local and distant correlations. We develop various models using different architectures, including U-Net, ViT, and Mamba, and different number of parameters. We evaluate the FLoating-point OPerations (FLOPs) needed by the models. We fine-tune on the PhilEO Bench for different downstream tasks: roads, buildings, and land cover. For most n-shots for roads and buildings, U-Net 200M-2T outperforms the other models. Using Mamba, we achieve comparable results on the downstream tasks, with less computational expenses. We also compare with the recent FM TerraMind which we evaluate on PhilEO Bench.

Nikolaos Dionelis, Casper Fibaek, Luke Camilleri et al.

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.

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.

Gaston Lenczner, Adrien Chan-Hon-Tong, Bertrand Le Saux et al.

We propose in this article to build up a collaboration between a deep neural network and a human in the loop to swiftly obtain accurate segmentation maps of remote sensing images. In a nutshell, the agent iteratively interacts with the network to correct its initially flawed predictions. Concretely, these interactions are annotations representing the semantic labels. Our methodological contribution is twofold. First, we propose two interactive learning schemes to integrate user inputs into deep neural networks. The first one concatenates the annotations with the other network's inputs. The second one uses the annotations as a sparse ground-truth to retrain the network. Second, we propose an active learning strategy to guide the user towards the most relevant areas to annotate. To this purpose, we compare different state-of-the-art acquisition functions to evaluate the neural network uncertainty such as ConfidNet, entropy or ODIN. Through experiments on three remote sensing datasets, we show the effectiveness of the proposed methods. Notably, we show that active learning based on uncertainty estimation enables to quickly lead the user towards mistakes and that it is thus relevant to guide the user interventions.

Rodrigo Caye Daudt, Bertrand Le Saux, Alexandre Boulch et al.

Large scale datasets created from crowdsourced labels or openly available data have become crucial to provide training data for large scale learning algorithms. While these datasets are easier to acquire, the data are frequently noisy and unreliable, which is motivating research on weakly supervised learning techniques. In this paper we propose original ideas that help us to leverage such datasets in the context of change detection. First, we propose the guided anisotropic diffusion (GAD) algorithm, which improves semantic segmentation results using the input images as guides to perform edge preserving filtering. We then show its potential in two weakly-supervised learning strategies tailored for change detection. The first strategy is an iterative learning method that combines model optimisation and data cleansing using GAD to extract the useful information from a large scale change detection dataset generated from open vector data. The second one incorporates GAD within a novel spatial attention layer that increases the accuracy of weakly supervised networks trained to perform pixel-level predictions from image-level labels. Improvements with respect to state-of-the-art are demonstrated on 4 different public datasets.

Christel Chappuis, Sylvain Lobry, Benjamin Kellenberger et al.

Visual question answering (VQA) has recently been introduced to remote sensing to make information extraction from overhead imagery more accessible to everyone. VQA considers a question (in natural language, therefore easy to formulate) about an image and aims at providing an answer through a model based on computer vision and natural language processing methods. As such, a VQA model needs to jointly consider visual and textual features, which is frequently done through a fusion step. In this work, we study three different fusion methodologies in the context of VQA for remote sensing and analyse the gains in accuracy with respect to the model complexity. Our findings indicate that more complex fusion mechanisms yield an improved performance, yet that seeking a trade-of between model complexity and performance is worthwhile in practice.

Alessandro Sebastianelli, Daniela A. Zaidenberg, Dario Spiller et al.

This article aims to investigate how circuit-based hybrid Quantum Convolutional Neural Networks (QCNNs) can be successfully employed as image classifiers in the context of remote sensing. The hybrid QCNNs enrich the classical architecture of CNNs by introducing a quantum layer within a standard neural network. The novel QCNN proposed in this work is applied to the Land Use and Land Cover (LULC) classification, chosen as an Earth Observation (EO) use case, and tested on the EuroSAT dataset used as reference benchmark. The results of the multiclass classification prove the effectiveness of the presented approach, by demonstrating that the QCNN performances are higher than the classical counterparts. Moreover, investigation of various quantum circuits shows that the ones exploiting quantum entanglement achieve the best classification scores. This study underlines the potentialities of applying quantum computing to an EO case study and provides the theoretical and experimental background for futures investigations.

Rémy Leroy, Pauline Trouvé-Peloux, Frédéric Champagnat et al.

Good quality reconstruction and comprehension of a scene rely on 3D estimation methods. The 3D information was usually obtained from images by stereo-photogrammetry, but deep learning has recently provided us with excellent results for monocular depth estimation. Building up a sufficiently large and rich training dataset to achieve these results requires onerous processing. In this paper, we address the problem of learning outdoor 3D point cloud from monocular data using a sparse ground-truth dataset. We propose Pix2Point, a deep learning-based approach for monocular 3D point cloud prediction, able to deal with complete and challenging outdoor scenes. Our method relies on a 2D-3D hybrid neural network architecture, and a supervised end-to-end minimisation of an optimal transport divergence between point clouds. We show that, when trained on sparse point clouds, our simple promising approach achieves a better coverage of 3D outdoor scenes than efficient monocular depth methods.

Veda Sunkara, Matthew Purri, Bertrand Le Saux et al.

To address the mounting destruction caused by floods in climate-vulnerable regions, we propose Street to Cloud, a machine learning pipeline for incorporating crowdsourced ground truth data into the segmentation of satellite imagery of floods. We propose this approach as a solution to the labor-intensive task of generating high-quality, hand-labeled training data, and demonstrate successes and failures of different plausible crowdsourcing approaches in our model. Street to Cloud leverages community reporting and machine learning to generate novel, near-real time insights into the extent of floods to be used for emergency response.

Javiera Castillo-Navarro, Bertrand Le Saux, Alexandre Boulch et al.

The development of semi-supervised learning techniques is essential to enhance the generalization capacities of machine learning algorithms. Indeed, raw image data are abundant while labels are scarce, therefore it is crucial to leverage unlabeled inputs to build better models. The availability of large databases have been key for the development of learning algorithms with high level performance. Despite the major role of machine learning in Earth Observation to derive products such as land cover maps, datasets in the field are still limited, either because of modest surface coverage, lack of variety of scenes or restricted classes to identify. We introduce a novel large-scale dataset for semi-supervised semantic segmentation in Earth Observation, the MiniFrance suite. MiniFrance has several unprecedented properties: it is large-scale, containing over 2000 very high resolution aerial images, accounting for more than 200 billions samples (pixels); it is varied, covering 16 conurbations in France, with various climates, different landscapes, and urban as well as countryside scenes; and it is challenging, considering land use classes with high-level semantics. Nevertheless, the most distinctive quality of MiniFrance is being the only dataset in the field especially designed for semi-supervised learning: it contains labeled and unlabeled images in its training partition, which reproduces a life-like scenario. Along with this dataset, we present tools for data representativeness analysis in terms of appearance similarity and a thorough study of MiniFrance data, demonstrating that it is suitable for learning and generalizes well in a semi-supervised setting. Finally, we present semi-supervised deep architectures based on multi-task learning and the first experiments on MiniFrance.

Gaston Lenczner, Adrien Chan-Hon-Tong, Nicola Luminari et al.

Dense pixel-wise classification maps output by deep neural networks are of extreme importance for scene understanding. However, these maps are often partially inaccurate due to a variety of possible factors. Therefore, we propose to interactively refine them within a framework named DISCA (Deep Image Segmentation with Continual Adaptation). It consists of continually adapting a neural network to a target image using an interactive learning process with sparse user annotations as ground-truth. We show through experiments on three datasets using synthesized annotations the benefits of the approach, reaching an IoU improvement up to 4.7% for ten sampled clicks. Finally, we exhibit that our approach can be particularly rewarding when it is faced to additional issues such as domain adaptation.

Marcela Carvalho, Maxime Ferrera, Alexandre Boulch et al.

This paper is a technical report about our submission for the ECCV 2018 3DRMS Workshop Challenge on Semantic 3D Reconstruction \cite{Tylecek2018rms}. In this paper, we address 3D semantic reconstruction for autonomous navigation using co-learning of depth map and semantic segmentation. The core of our pipeline is a deep multi-task neural network which tightly refines depth and also produces accurate semantic segmentation maps. Its inputs are an image and a raw depth map produced from a pair of images by standard stereo vision. The resulting semantic 3D point clouds are then merged in order to create a consistent 3D mesh, in turn used to produce dense semantic 3D reconstruction maps. The performances of each step of the proposed method are evaluated on the dataset and multiple tasks of the 3DRMS Challenge, and repeatedly surpass state-of-the-art approaches.

Nicolas Audebert, Alexandre Boulch, Bertrand Le Saux et al.

Understanding visual scenes relies more and more on dense pixel-wise classification obtained via deep fully convolutional neural networks. However, due to the nature of the networks, predictions often suffer from blurry boundaries and ill-segmented shapes, fueling the need for post-processing. This work introduces a new semantic segmentation regularization based on the regression of a distance transform. After computing the distance transform on the label masks, we train a FCN in a multi-task setting in both discrete and continuous spaces by learning jointly classification and distance regression. This requires almost no modification of the network structure and adds a very low overhead to the training process. Learning to approximate the distance transform back-propagates spatial cues that implicitly regularizes the segmentation. We validate this technique with several architectures on various datasets, and we show significant improvements compared to competitive baselines.

Rodrigo Caye Daudt, Bertrand Le Saux, Alexandre Boulch et al.

Large scale datasets created from user labels or openly available data have become crucial to provide training data for large scale learning algorithms. While these datasets are easier to acquire, the data are frequently noisy and unreliable, which is motivating research on weakly supervised learning techniques. In this paper we propose an iterative learning method that extracts the useful information from a large scale change detection dataset generated from open vector data to train a fully convolutional network which surpasses the performance obtained by naive supervised learning. We also propose the guided anisotropic diffusion algorithm, which improves semantic segmentation results using the input images as guides to perform edge preserving filtering, and is used in conjunction with the iterative training method to improve results.

Rodrigo Caye Daudt, Bertrand Le Saux, Alexandre Boulch et al.

The Copernicus Sentinel-2 program now provides multispectral images at a global scale with a high revisit rate. In this paper we explore the usage of convolutional neural networks for urban change detection using such multispectral images. We first present the new change detection dataset that was used for training the proposed networks, which will be openly available to serve as a benchmark. The Onera Satellite Change Detection (OSCD) dataset is composed of pairs of multispectral aerial images, and the changes were manually annotated at pixel level. We then propose two architectures to detect changes, Siamese and Early Fusion, and compare the impact of using different numbers of spectral channels as inputs. These architectures are trained from scratch using the provided dataset.

Rodrigo Caye Daudt, Bertrand Le Saux, Alexandre Boulch

This paper presents three fully convolutional neural network architectures which perform change detection using a pair of coregistered images. Most notably, we propose two Siamese extensions of fully convolutional networks which use heuristics about the current problem to achieve the best results in our tests on two open change detection datasets, using both RGB and multispectral images. We show that our system is able to learn from scratch using annotated change detection images. Our architectures achieve better performance than previously proposed methods, while being at least 500 times faster than related systems. This work is a step towards efficient processing of data from large scale Earth observation systems such as Copernicus or Landsat.

Rodrigo Caye Daudt, Bertrand Le Saux, Alexandre Boulch et al.

Change detection is one of the main problems in remote sensing, and is essential to the accurate processing and understanding of the large scale Earth observation data available through programs such as Sentinel and Landsat. Most of the recently proposed change detection methods bring deep learning to this context, but openly available change detection datasets are still very scarce, which limits the methods that can be proposed and tested. In this paper we present the first large scale high resolution semantic change detection (HRSCD) dataset, which enables the usage of deep learning methods for semantic change detection. The dataset contains coregistered RGB image pairs, pixel-wise change information and land cover information. We then propose several methods using fully convolutional neural networks to perform semantic change detection. Most notably, we present a network architecture that performs change detection and land cover mapping simultaneously, while using the predicted land cover information to help to predict changes. We also describe a sequential training scheme that allows this network to be trained without setting a hyperparameter that balances different loss functions and achieves the best overall results.

Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre

This work addresses the scarcity of annotated hyperspectral data required to train deep neural networks. Especially, we investigate generative adversarial networks and their application to the synthesis of consistent labeled spectra. By training such networks on public datasets, we show that these models are not only able to capture the underlying distribution, but also to generate genuine-looking and physically plausible spectra. Moreover, we experimentally validate that the synthetic samples can be used as an effective data augmentation strategy. We validate our approach on several public hyper-spectral datasets using a variety of deep classifiers.

Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre

In this work, we investigate various methods to deal with semantic labeling of very high resolution multi-modal remote sensing data. Especially, we study how deep fully convolutional networks can be adapted to deal with multi-modal and multi-scale remote sensing data for semantic labeling. Our contributions are threefold: a) we present an efficient multi-scale approach to leverage both a large spatial context and the high resolution data, b) we investigate early and late fusion of Lidar and multispectral data, c) we validate our methods on two public datasets with state-of-the-art results. Our results indicate that late fusion make it possible to recover errors steaming from ambiguous data, while early fusion allows for better joint-feature learning but at the cost of higher sensitivity to missing data.

Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre

In this work, we investigate the use of OpenStreetMap data for semantic labeling of Earth Observation images. Deep neural networks have been used in the past for remote sensing data classification from various sensors, including multispectral, hyperspectral, SAR and LiDAR data. While OpenStreetMap has already been used as ground truth data for training such networks, this abundant data source remains rarely exploited as an input information layer. In this paper, we study different use cases and deep network architectures to leverage OpenStreetMap data for semantic labeling of aerial and satellite images. Especially , we look into fusion based architectures and coarse-to-fine segmentation to include the OpenStreetMap layer into multispectral-based deep fully convolutional networks. We illustrate how these methods can be successfully used on two public datasets: ISPRS Potsdam and DFC2017. We show that OpenStreetMap data can efficiently be integrated into the vision-based deep learning models and that it significantly improves both the accuracy performance and the convergence speed of the networks.

Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre

In this work, we present a novel module to perform fusion of heterogeneous data using fully convolutional networks for semantic labeling. We introduce residual correction as a way to learn how to fuse predictions coming out of a dual stream architecture. Especially, we perform fusion of DSM and IRRG optical data on the ISPRS Vaihingen dataset over a urban area and obtain new state-of-the-art results.

Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre

In this paper, we investigate the impact of segmentation algorithms as a preprocessing step for classification of remote sensing images in a deep learning framework. Especially, we address the issue of segmenting the image into regions to be classified using pre-trained deep neural networks as feature extractors for an SVM-based classifier. An efficient segmentation as a preprocessing step helps learning by adding a spatially-coherent structure to the data. Therefore, we compare algorithms producing superpixels with more traditional remote sensing segmentation algorithms and measure the variation in terms of classification accuracy. We establish that superpixel algorithms allow for a better classification accuracy as a homogenous and compact segmentation favors better generalization of the training samples.