Ksenia Bittner
YOU?
Author Swipe
View article: AIM2PC: Aerial Image to 3D Building Point Cloud Reconstruction
AIM2PC: Aerial Image to 3D Building Point Cloud Reconstruction Open
Three-dimensional urban reconstruction of buildings from single-view images has attracted significant attention over the past two decades. However, recent methods primarily focus on rooftops from aerial images, often overlooking essential …
View article: Efficient Building Roof Type Classification: A Domain-Specific Self-Supervised Approach
Efficient Building Roof Type Classification: A Domain-Specific Self-Supervised Approach Open
Accurate classification of building roof types from aerial imagery is crucial for various remote sensing applications, including urban planning, disaster management, and infrastructure monitoring. However, this task is often hindered by th…
View article: Building Segmentation and Modelling from Space-Borne and Aerial Imagery
Building Segmentation and Modelling from Space-Borne and Aerial Imagery Open
Accurate 3D building reconstruction is essential for urban planning, disaster management, and environmental applications. However, current methods often struggle to achieve geometric precision and topological consistency, particularly when…
View article: Can Location Embeddings Enhance Super-Resolution of Satellite Imagery?
Can Location Embeddings Enhance Super-Resolution of Satellite Imagery? Open
Publicly available satellite imagery, such as Sentinel- 2, often lacks the spatial resolution required for accurate analysis of remote sensing tasks including urban planning and disaster response. Current super-resolution techniques are ty…
View article: Dfilled: Repurposing Edge-Enhancing Diffusion for Guided DSM Void Filling
Dfilled: Repurposing Edge-Enhancing Diffusion for Guided DSM Void Filling Open
Digital Surface Models (DSMs) are essential for accurately representing Earth's topography in geospatial analyses. DSMs capture detailed elevations of natural and manmade features, crucial for applications like urban planning, vegetation s…
View article: Multi-branch convolutional neural network in building polygonization using remote sensing images
Multi-branch convolutional neural network in building polygonization using remote sensing images Open
Building extraction and polygonization is important for urban studies, such as urbanization monitoring, urban planning. Remote sensing images, especially in RGB bands, provide sufficient semantic information which is useful for the task of…
View article: Roof plane parsing towards LoD-2.2 building reconstruction based on joint learning using remote sensing images
Roof plane parsing towards LoD-2.2 building reconstruction based on joint learning using remote sensing images Open
Building models are important for urban studies. Remote sensing multi-spectral (MS) images are widely used for its rich semantic information. The lack of geometry features is fulfilled by introducing photogrammetry derived digital surface …
View article: Rectilinear Building Footprint Regularization Using Deep Learning
Rectilinear Building Footprint Regularization Using Deep Learning Open
Nowadays, deep learning allows to automatically learn features from data. Buildings are one of the most important objects in urban environments. They are used in applications such as inputs to building reconstruction, disaster monitoring, …
View article: Unit-level LoD2 Building Reconstruction from Satellite-derived Digital Surface Model and Orthophoto
Unit-level LoD2 Building Reconstruction from Satellite-derived Digital Surface Model and Orthophoto Open
Recent advancements in deep learning have enabled the possibility to identify unit-level building sections from very high resolution satellite images. By learning from the examples, deep models can capture patterns from the low-resolution …
View article: Real-GDSR: Real-World Guided DSM Super-Resolution via Edge-Enhancing Residual Network
Real-GDSR: Real-World Guided DSM Super-Resolution via Edge-Enhancing Residual Network Open
A low-resolution digital surface model (DSM) features distinctive attributes impacted by noise, sensor limitations and data acquisition conditions, which failed to be replicated using simple interpolation methods like bicubic. This causes …
View article: PLANES4LOD2: Reconstruction of LoD-2 building models using a depth attention-based fully convolutional neural network
PLANES4LOD2: Reconstruction of LoD-2 building models using a depth attention-based fully convolutional neural network Open
Level of detail (LoD)-2 reconstruction is an inevitable task in digital twin-related applications such as disaster management, flood simulation, landslide simulation and solar panel recommendation. However, there is a lack of capable metho…
View article: SyntStereo2Real: Edge-Aware GAN for Remote Sensing Image-to-Image Translation while Maintaining Stereo Constraint
SyntStereo2Real: Edge-Aware GAN for Remote Sensing Image-to-Image Translation while Maintaining Stereo Constraint Open
In the field of remote sensing, the scarcity of stereo-matched and particularly lack of accurate ground truth data often hinders the training of deep neural networks. The use of synthetically generated images as an alternative, alleviates …
View article: Real-GDSR: Real-World Guided DSM Super-Resolution via Edge-Enhancing Residual Network
Real-GDSR: Real-World Guided DSM Super-Resolution via Edge-Enhancing Residual Network Open
A low-resolution digital surface model (DSM) features distinctive attributes impacted by noise, sensor limitations and data acquisition conditions, which failed to be replicated using simple interpolation methods like bicubic. This causes …
View article: EVALUATING CONVNET AND TRANSFORMER BASED SELF-SUPERVISED ALGORITHMS FOR BUILDING ROOF FORM CLASSIFICATION
EVALUATING CONVNET AND TRANSFORMER BASED SELF-SUPERVISED ALGORITHMS FOR BUILDING ROOF FORM CLASSIFICATION Open
This research paper presents a comprehensive evaluation of various self-supervised learning models for building roof type classification. We conduct linear evaluation experiments for the models pretrained on both the ImageNet1K dataset and…
View article: DSM2DTM: AN END-TO-END DEEP LEARNING APPROACH FOR DIGITAL TERRAIN MODEL GENERATION
DSM2DTM: AN END-TO-END DEEP LEARNING APPROACH FOR DIGITAL TERRAIN MODEL GENERATION Open
Remotely sensed Earth elevation data or digital surface model (DSM) typically contains both terrain and above-ground information such as vegetation and man-made constructions. However, many applications require pure bare-terrain data, also…
View article: ROOF3D: A REAL AND SYNTHETIC DATA COLLECTION FOR INDIVIDUAL BUILDING ROOF PLANE AND BUILDING SECTIONS DETECTION
ROOF3D: A REAL AND SYNTHETIC DATA COLLECTION FOR INDIVIDUAL BUILDING ROOF PLANE AND BUILDING SECTIONS DETECTION Open
Deep learning is a powerful tool to extract both individual building and roof plane polygons. But deep learning requires a large amount of labeled data. Hence, publicly available level of detail (LoD)-2 datasets are a natural choice to tra…
View article: Deep Learning for the Automatic Division of Building Constructions Into Sections on Remote Sensing Images
Deep Learning for the Automatic Division of Building Constructions Into Sections on Remote Sensing Images Open
Urban areas predominantly consist of complex building structures, which are assembled of multiple building sections. From very high resolution remote sensing imagery, not only roof-tops but also the separation lines between them are visibl…
View article: BUILDING SECTION INSTANCE SEGMENTATION WITH COMBINED CLASSICAL AND DEEP LEARNING METHODS
BUILDING SECTION INSTANCE SEGMENTATION WITH COMBINED CLASSICAL AND DEEP LEARNING METHODS Open
In big cities, the complexity of urban infrastructure is very high. In city centers, one construction can consist of several building sections of different heights or roof geometries. Most of the existing approaches detect those buildings …
View article: Machine-learned 3D Building Vectorization from Satellite Imagery
Machine-learned 3D Building Vectorization from Satellite Imagery Open
We propose a machine learning based approach for automatic 3D building reconstruction and vectorization. Taking a single-channel photogrammetric digital surface model (DSM) and panchromatic (PAN) image as input, we first filter out non-bui…
View article: Machine-learned Regularization and Polygonization of Building Segmentation Masks
Machine-learned Regularization and Polygonization of Building Segmentation Masks Open
We propose a machine learning based approach for automatic regularization and polygonization of building segmentation masks. Taking an image as input, we first predict building segmentation maps exploiting generic fully convolutional netwo…
View article: Stepwise Refinement Of Low Resolution Labels For Earth Observation Data: Part 2
Stepwise Refinement Of Low Resolution Labels For Earth Observation Data: Part 2 Open
This paper describes the contribution of the DLR team ranking 2nd in Track 2 of the 2020 IEEE GRSS Data Fusion Contest. The semantic classification of multimodal earth observation data proposed is based on the refinement of low resolution …
View article: Stepwise Refinement Of Low Resolution Labels For Earth Observation Data: Part 1
Stepwise Refinement Of Low Resolution Labels For Earth Observation Data: Part 1 Open
This paper describes the contribution of the DLR team
\nranking 3rd in Track 1 of the 2020 IEEE GRSS Data Fusion
\nContest, with results ranking 2nd in Track 2 of the same contest are reported in a companion paper. The classifications
\nar…
View article: Map-Repair: Deep Cadastre Maps Alignment and Temporal Inconsistencies Fix in Satellite Images
Map-Repair: Deep Cadastre Maps Alignment and Temporal Inconsistencies Fix in Satellite Images Open
In the fast developing countries it is hard to trace new buildings construction or old structures destruction and, as a result, to keep the up-to-date cadastre maps. Moreover, due to the complexity of urban regions or inconsistency of data…
View article: LONG-SHORT SKIP CONNECTIONS IN DEEP NEURAL NETWORKS FOR DSM REFINEMENT
LONG-SHORT SKIP CONNECTIONS IN DEEP NEURAL NETWORKS FOR DSM REFINEMENT Open
Detailed digital surface models (DSMs) from space-borne sensors are the key to successful solutions for many remote sensing problems, like environmental disaster simulations, change detection in rural and urban areas, 3D urban modeling for…
View article: Map-Repair: Deep Cadastre Maps Alignment and Temporal Inconsistencies\n Fix in Satellite Images
Map-Repair: Deep Cadastre Maps Alignment and Temporal Inconsistencies\n Fix in Satellite Images Open
In the fast developing countries it is hard to trace new buildings\nconstruction or old structures destruction and, as a result, to keep the\nup-to-date cadastre maps. Moreover, due to the complexity of urban regions or\ninconsistency of d…
View article: A generalized multi-task learning approach to stereo DSM filtering in urban areas
A generalized multi-task learning approach to stereo DSM filtering in urban areas Open
View article: A Generalized Multi-Task Learning Approach to Stereo DSM Filtering in\n Urban Areas
A Generalized Multi-Task Learning Approach to Stereo DSM Filtering in\n Urban Areas Open
City models and height maps of urban areas serve as a valuable data source\nfor numerous applications, such as disaster management or city planning. While\nthis information is not globally available, it can be substituted by digital\nsurfa…
View article: DSM Building Shape Refinement from Combined Remote Sensing Images Based on WNET-CGANS
DSM Building Shape Refinement from Combined Remote Sensing Images Based on WNET-CGANS Open
We describe the workflow of a digital surface models (DSMs) refinement algorithm using a hybrid conditional generative adversarial network (cGAN) where the generative part consists of two parallel networks merged at the last stage forming …
View article: Multi-Task cGAN for Simultaneous Spaceborne DSM Refinement and Roof-Type Classification
Multi-Task cGAN for Simultaneous Spaceborne DSM Refinement and Roof-Type Classification Open
Various deep learning applications benefit from multi-task learning with multiple regression and classification objectives by taking advantage of the similarities between individual tasks. This can result in improved learning efficiency an…
View article: Late or Earlier Information Fusion from Depth and Spectral Data? Large-Scale Digital Surface Model Refinement by Hybrid-cGAN
Late or Earlier Information Fusion from Depth and Spectral Data? Large-Scale Digital Surface Model Refinement by Hybrid-cGAN Open
We present the workflow of a DSM refinement methodology using a Hybrid-cGAN where the generative part consists of two encoders and a common decoder which blends the spectral and height information within one network. The inputs to the Hybr…