Searcharxiv⌕ Search

arXiv subjects

Matthias Braun

Publications and source records attributed to Matthias Braun.

At least 19 recordsLinked to original sources
arXiv2025

AMD-HookNet++: Evolution of AMD-HookNet with Hybrid CNN-Transformer Feature Enhancement for Glacier Calving Front Segmentation

The dynamics of glaciers and ice shelf fronts significantly impact the mass balance of ice sheets and coastal sea levels. To effectively monitor glacier conditions, it is crucial to consistently estimate positional shifts of glacier calving fronts. AMD-HookNet firstly introduces a pure two-branch convolutional neural network (CNN) for glacier segmentation. Yet, the local nature and translational invariance of convolution operations, while beneficial for capturing low-level details, restricts the model ability to maintain long-range dependencies. In this study, we propose AMD-HookNet++, a novel advanced hybrid CNN-Transformer feature enhancement method for segmenting glaciers and delineating calving fronts in synthetic aperture radar images. Our hybrid structure consists of two branches: a Transformer-based context branch to capture long-range dependencies, which provides global contextual information in a larger view, and a CNN-based target branch to preserve local details. To strengthen the representation of the connected hybrid features, we devise an enhanced spatial-channel attention module to foster interactions between the hybrid CNN-Transformer branches through dynamically adjusting the token relationships from both spatial and channel perspectives. Additionally, we develop a pixel-to-pixel contrastive deep supervision to optimize our hybrid model by integrating pixelwise metric learning into glacier segmentation. Through extensive experiments and comprehensive quantitative and qualitative analyses on the challenging glacier segmentation benchmark dataset CaFFe, we show that AMD-HookNet++ sets a new state of the art with an IoU of 78.2 and a HD95 of 1,318 m, while maintaining a competitive MDE of 367 m. More importantly, our hybrid model produces smoother delineations of calving fronts, resolving the issue of jagged edges typically seen in pure Transformer-based approaches.

cs.CV↗
arXiv2025

Multi-temporal Calving Front Segmentation

The calving fronts of marine-terminating glaciers undergo constant changes. These changes significantly affect the glacier's mass and dynamics, demanding continuous monitoring. To address this need, deep learning models were developed that can automatically delineate the calving front in Synthetic Aperture Radar imagery. However, these models often struggle to correctly classify areas affected by seasonal conditions such as ice melange or snow-covered surfaces. To address this issue, we propose to process multiple frames from a satellite image time series of the same glacier in parallel and exchange temporal information between the corresponding feature maps to stabilize each prediction. We integrate our approach into the current state-of-the-art architecture Tyrion and accomplish a new state-of-the-art performance on the CaFFe benchmark dataset. In particular, we achieve a Mean Distance Error of 184.4 m and a mean Intersection over Union of 83.6.

cs.CV↗
arXiv2025

SSL4SAR: Self-Supervised Learning for Glacier Calving Front Extraction from SAR Imagery

Glaciers are losing ice mass at unprecedented rates, increasing the need for accurate, year-round monitoring to understand frontal ablation, particularly the factors driving the calving process. Deep learning models can extract calving front positions from Synthetic Aperture Radar imagery to track seasonal ice losses at the calving fronts of marine- and lake-terminating glaciers. The current state-of-the-art model relies on ImageNet-pretrained weights. However, they are suboptimal due to the domain shift between the natural images in ImageNet and the specialized characteristics of remote sensing imagery, in particular for Synthetic Aperture Radar imagery. To address this challenge, we propose two novel self-supervised multimodal pretraining techniques that leverage SSL4SAR, a new unlabeled dataset comprising 9,563 Sentinel-1 and 14 Sentinel-2 images of Arctic glaciers, with one optical image per glacier in the dataset. Additionally, we introduce a novel hybrid model architecture that combines a Swin Transformer encoder with a residual Convolutional Neural Network (CNN) decoder. When pretrained on SSL4SAR, this model achieves a mean distance error of 293 m on the "CAlving Fronts and where to Find thEm" (CaFFe) benchmark dataset, outperforming the prior best model by 67 m. Evaluating an ensemble of the proposed model on a multi-annotator study of the benchmark dataset reveals a mean distance error of 75 m, approaching the human performance of 38 m. This advancement enables precise monitoring of seasonal changes in glacier calving fronts.

cs.CV↗
arXiv2023

Foundational Competencies and Responsibilities of a Research Software Engineer: Current State and Suggestions for Future Directions

The term Research Software Engineer, or RSE, emerged a little over 10 years ago as a way to represent individuals working in the research community but focusing on software development. The term has been widely adopted and there are a number of high-level definitions of what an RSE is. However, the roles of RSEs vary depending on the institutional context they work in. At one end of the spectrum, RSE roles may look similar to a traditional research role. At the other extreme, they resemble that of a software engineer in industry. Most RSE roles inhabit the space between these two extremes. Therefore, providing a straightforward, comprehensive definition of what an RSE does and what experience, skills and competencies are required to become one is challenging. In this community paper we define the broad notion of what an RSE is, explore the different types of work they undertake, and define a list of fundamental competencies as well as values that define the general profile of an RSE. On this basis, we elaborate on the progression of these skills along different dimensions, looking at specific types of RSE roles, proposing recommendations for organisations, and giving examples of future specialisations. An appendix details how existing curricula fit into this framework.

cs.SE↗
arXiv2023

AMD-HookNet for Glacier Front Segmentation

Knowledge on changes in glacier calving front positions is important for assessing the status of glaciers. Remote sensing imagery provides the ideal database for monitoring calving front positions, however, it is not feasible to perform this task manually for all calving glaciers globally due to time-constraints. Deep learning-based methods have shown great potential for glacier calving front delineation from optical and radar satellite imagery. The calving front is represented as a single thin line between the ocean and the glacier, which makes the task vulnerable to inaccurate predictions. The limited availability of annotated glacier imagery leads to a lack of data diversity (not all possible combinations of different weather conditions, terminus shapes, sensors, etc. are present in the data), which exacerbates the difficulty of accurate segmentation. In this paper, we propose Attention-Multi-hooking-Deep-supervision HookNet (AMD-HookNet), a novel glacier calving front segmentation framework for synthetic aperture radar (SAR) images. The proposed method aims to enhance the feature representation capability through multiple information interactions between low-resolution and high-resolution inputs based on a two-branch U-Net. The attention mechanism, integrated into the two branch U-Net, aims to interact between the corresponding coarse and fine-grained feature maps. This allows the network to automatically adjust feature relationships, resulting in accurate pixel-classification predictions. Extensive experiments and comparisons on the challenging glacier segmentation benchmark dataset CaFFe show that our AMD-HookNet achieves a mean distance error of 438 m to the ground truth outperforming the current state of the art by 42%, which validates its effectiveness.

cs.CV↗
arXiv2021

Bayesian U-Net for Segmenting Glaciers in SAR Imagery

Fluctuations of the glacier calving front have an important influence over the ice flow of whole glacier systems. It is therefore important to precisely monitor the position of the calving front. However, the manual delineation of SAR images is a difficult, laborious and subjective task. Convolutional neural networks have previously shown promising results in automating the glacier segmentation in SAR images, making them desirable for further exploration of their possibilities. In this work, we propose to compute uncertainty and use it in an Uncertainty Optimization regime as a novel two-stage process. By using dropout as a random sampling layer in a U-Net architecture, we create a probabilistic Bayesian Neural Network. With several forward passes, we create a sampling distribution, which can estimate the model uncertainty for each pixel in the segmentation mask. The additional uncertainty map information can serve as a guideline for the experts in the manual annotation of the data. Furthermore, feeding the uncertainty map to the network leads to 95.24% Dice similarity, which is an overall improvement in the segmentation performance compared to the state-of-the-art deterministic U-Net-based glacier segmentation pipelines.

cs.LG↗
arXiv2021

On Mathews Correlation Coefficient and Improved Distance Map Loss for Automatic Glacier Calving Front Segmentation in SAR Imagery

The vast majority of the outlet glaciers and ice streams of the polar ice sheets end in the ocean. Ice mass loss via calving of the glaciers into the ocean has increased over the last few decades. Information on the temporal variability of the calving front position provides fundamental information on the state of the glacier and ice stream, which can be exploited as calibration and validation data to enhance ice dynamics modeling. To identify the calving front position automatically, deep neural network-based semantic segmentation pipelines can be used to delineate the acquired SAR imagery. However, the extreme class imbalance is highly challenging for the accurate calving front segmentation in these images. Therefore, we propose the use of the Mathews correlation coefficient (MCC) as an early stopping criterion because of its symmetrical properties and its invariance towards class imbalance. Moreover, we propose an improvement to the distance map-based binary cross-entropy (BCE) loss function. The distance map adds context to the loss function about the important regions for segmentation and helps accounting for the imbalanced data. Using Mathews correlation coefficient as early stopping demonstrates an average 15% dice coefficient improvement compared to the commonly used BCE. The modified distance map loss further improves the segmentation performance by another 2%. These results are encouraging as they support the effectiveness of the proposed methods for segmentation problems suffering from extreme class imbalances.

eess.IV↗
arXiv2021

Pixel-wise Distance Regression for Glacier Calving Front Detection and Segmentation

Glacier calving front position (CFP) is an important glaciological variable. Traditionally, delineating the CFPs has been carried out manually, which was subjective, tedious and expensive. Automating this process is crucial for continuously monitoring the evolution and status of glaciers. Recently, deep learning approaches have been investigated for this application. However, the current methods get challenged by a severe class-imbalance problem. In this work, we propose to mitigate the class-imbalance between the calving front class and the non-calving front class by reformulating the segmentation problem into a pixel-wise regression task. A Convolutional Neural Network gets optimized to predict the distance values to the glacier front for each pixel in the image. The resulting distance map localizes the CFP and is further post-processed to extract the calving front line. We propose three post-processing methods, one method based on statistical thresholding, a second method based on conditional random fields (CRF), and finally the use of a second U-Net. The experimental results confirm that our approach significantly outperforms the state-of-the-art methods and produces accurate delineation. The Second U-Net obtains the best performance results, resulting in an average improvement of about 21% dice coefficient enhancement.

cs.CV↗
arXiv2021

Synthetic Glacier SAR Image Generation from Arbitrary Masks Using Pix2Pix Algorithm

Supervised machine learning requires a large amount of labeled data to achieve proper test results. However, generating accurately labeled segmentation maps on remote sensing imagery, including images from synthetic aperture radar (SAR), is tedious and highly subjective. In this work, we propose to alleviate the issue of limited training data by generating synthetic SAR images with the pix2pix algorithm. This algorithm uses conditional Generative Adversarial Networks (cGANs) to generate an artificial image while preserving the structure of the input. In our case, the input is a segmentation mask, from which a corresponding synthetic SAR image is generated. We present different models, perform a comparative study and demonstrate that this approach synthesizes convincing glaciers in SAR images with promising qualitative and quantitative results.

cs.LG↗
arXiv2021

Glacier Calving Front Segmentation Using Attention U-Net

An essential climate variable to determine the tidewater glacier status is the location of the calving front position and the separation of seasonal variability from long-term trends. Previous studies have proposed deep learning-based methods to semi-automatically delineate the calving fronts of tidewater glaciers. They used U-Net to segment the ice and non-ice regions and extracted the calving fronts in a post-processing step. In this work, we show a method to segment the glacier calving fronts from SAR images in an end-to-end fashion using Attention U-Net. The main objective is to investigate the attention mechanism in this application. Adding attention modules to the state-of-the-art U-Net network lets us analyze the learning process by extracting its attention maps. We use these maps as a tool to search for proper hyperparameters and loss functions in order to generate higher qualitative results. Our proposed attention U-Net performs comparably to the standard U-Net while providing additional insight into those regions on which the network learned to focus more. In the best case, the attention U-Net achieves a 1.5% better Dice score compared to the canonical U-Net with a glacier front line prediction certainty of up to 237.12 meters.

cs.LG↗
arXiv2018

Kähler forms for families of Calabi-Yau manifolds

Kähler-Einstein metrics for polarized families of Calabi-Yau manifolds define a natural hermitian metric on the relative canonical bundle. The fact that the curvature form is equal to the pull-back of the Weil-Petersson form up to a numerical constant is being used for the construction of a Kähler form on the total space of a given family, whose restriction to the fibers is Ricci flat.

math.CV↗
arXiv2018

Positivity of direct images of fiberwise Ricci-flat metrics on Calabi-Yau fibrations

Let $X$ be a Kähler manifold which is fibered over a complex manifold $Y$ such that every fiber is a Calabi-Yau manifold. Let $ω$ be a fixed Kähler form on $X$. By Yau's theorem, there exists a unique Ricci-flat Kähler form $ρ\vert_{X_y}$ for each fiber, which is cohomologous to $ω\vert_{X_y}$. This family of Ricci-flat Kähler forms $ρ\vert_{X_y}$ induces a smooth $(1,1)$-form $ρ$ on $X$ with a normalization condition. In this paper, we prove that the direct image of $ρ^{n+1}$ is positive on the base $Y$. We also discuss several byproducts, among them the local triviality of families of Calabi-Yau manifolds.

math.CV↗
arXiv2010

Spin exchange interaction with tunable range between graphene quantum dots

We study the spin exchange between two electrons localized in separate quantum dots in graphene. The electronic states in the conduction band are coupled indirectly by tunneling to a common continuum of delocalized states in the valence band. As a model, we use a two-impurity Anderson Hamiltonian which we subsequently transform into an effective spin Hamiltonian by way of a two-stage Schrieffer-Wolff transformation. We then compare our result to that from a Coqblin-Schrieffer approach as well as to fourth order perturbation theory.

cond-mat.mes-hall↗
arXiv2008

Signature of chirality in scanning-probe imaging of charge flow in graphene

We theoretically propose to directly observe the chiral nature of charge carriers in graphene mono- and bilayers within a controlled scattering experiment. The charge located on a capacitively coupled scanning probe microscope (SPM) tip acts as a scattering center with controllable position on the graphene sheet. Unambiguous features from the chirality of the particles in single and bilayer graphene arise in the ballistic transport in the presence of such a scattering center. To theoretically model the scattering from the smooth potential created by the SPM tip, we develop the scattering formalism within first-order Born approximation. We calculate the current through a device with an SPM tip between two constrictions (quantum point contacts) as a function of the tip position.

cond-mat.mes-hall↗
arXiv2008

Non-adiabatic two-parameter charge and spin pumping in a quantum dot

We study DC charge and spin transport through a weakly coupled quantum dot, driven by a non-adiabatic periodic change of system parameters. We generalize the model of Tien and Gordon to simultaneously oscillating voltages and tunnel couplings. When applying our general result to the two-parameter charge pumping in quantum dots, we find interference effects between the oscillations of the voltage and tunnel couplings. Furthermore, we discuss the possibility to electrically pump a spin current in presence of a static magnetic field.

cond-mat.mes-hall↗
arXiv2007

Theory of a Magnetically-Controlled Quantum-Dot Spin Transistor

We examine transport through a quantum dot coupled to three ferromagnetic leads in the regime of weak tunnel coupling. A finite source-drain voltage generates a nonequilibrium spin on the otherwise non-magnetic quantum dot. This spin accumulation leads to magnetoresistance. A ferromagnetic but current-free base electrode influences the quantum-dot spin via incoherent spin-flip processes and coherent spin precession. As the dot spin determines the conductance of the device, this allows for a purely magnetic transistor-like operation. We analyze the effect of both types of processes on the electric current in different geometries.

cond-mat.mes-hall↗
arXiv2006

Faraday-rotation fluctuation spectroscopy with static and oscillating magnetic fields

By Faraday-rotation fluctuation spectroscopy one measures the spin noise via Faraday-induced fluctuations of the polarization plane of a laser transmitting the sample. In the fist part of this paper, we present a theoretical model of recent experiments on alkali gas vapors and semiconductors, done in the presence of a {\em static} magnetic field. In a static field, the spin noise shows a resonance line, revealing the Larmor frequency and the spin coherence time $T_2$ of the electrons. Second, we discuss the possibility to use an {\em oscillating} magnetic field in the Faraday setup. With an oscillating field applied, one can observe multi-photon absorption processes in the spin noise. Furthermore an oscillating field could also help to avoid line broadening due to structural or chemical inhomogeneities in the sample, and thereby increase the precision of the spin-coherence time measurement.

cond-mat.mes-hall↗