Recent publications

In this work, we present an artificial intelligence (AI)-based semantic segmentation approach for three-dimensional (3D) point clouds which were generated from 2D images with a structure from motion (SfM) pipeline. We utilize state-of-the-art...

In this work, we propose FEDAUX, an extension to Federated Distillation, which, under the same set of assumptions, drastically improves the performance by deriving maximum utility from the unlabeled auxiliary data. Our proposed method achieves...

A lab-on-a-chip multichannel sensing platform for biomedical analysis based on optical silicon nitride (SiNx) microring-resonators (MRR) was established. The resonators were surface functionalized and finally combined with a microfluidic chamber...

This paper validates an in-house three-dimensional ray-launching (3D-RL) algorithm with a channel sounder measurement campaign that has been performed in a typical indoor environment at 300 GHz.

We present a hybrid photonic integrated mmW/THz signal source, which comprises two tunable lasers and on-chip wavelength meters. The continuous wavelength tunability of a single laser is over 12 nm (1.5 THz), and the wavelength meter accuracy is...

A novel polymer-based 1x4 vertical multimode interference (MMI) coupler for 3D photonics is presented. It connects four vertically stacked waveguide layers with a spacing of 21.6 µm. The functionality is demonstrated on a fabricated device.

Ultra-broadband wavelength conversion is identified as one of the key issues in future high capacity, flexible optical networks. In this contribution, methods to optimize the design of a multi-modal high-nonlinear SOI waveguide to achieve...

A lab-on-a-chip multichannel sensing platform for biomedical analysis based on optical silicon nitride (SiNx) mi- croring-resonators (MRR) was established. The resonators were surface functionalized and finally combined with a microfluidic...

Digitalization of existing buildings and the creation of 3D BIM models is crucial for many tasks. Of particular importance are floor plans, which contain information about building layouts and are vital for construction, maintenance or...

In this work, we enhance a professional end-to-end volumetric video production pipeline to achieve high-fidelity human body reconstruction using only passive cameras.We introduce and optimize deep learning based multi-view stereo networks for...

In this paper, we propose a template-free method to reconstruct surface geometry and appearance using neural implicit representations from multi-view videos. We leverage topology-aware deformation and the signed distance field to learn complex...

Oliver Schreer, Peter Eisert, Ingo Feldmann, Anna Hilsmann, Sylvain Renault, Marcus Zepp, Wieland Morgenstern, Rodrigo Mauricio Diaz Fernandez, Markus Worchel

Using wavelength conversion of our fabricated SOI strip waveguide, we compared experimentally the polarization-insensitive configuration toward S-band real-time transmission. It is found that parallel configuration is 3dB superior in in-out...

We propose an analysis-by-synthesis method for fast multi-view 3D reconstruction of opaque objects with arbitrary materials and illumination. We represent surfaces as triangle meshes and build a differentiable rendering pipeline around triangle...

The goal of pollution forecasting models is to allow the prediction and control of the air quality. While such deep learning models were deemed for a long time as black boxes, recent advances in eXplainable AI (XAI) allow to look through the...

Rule-based and modified backpropagation XAI methods struggle with innovative layer building blocks and implementation-invariance issues. 

In this work we propose canonizations for popular deep neural network architectures and...

We experimentally show modulation format identification in the optical domain using Silicon-on-Insulator-based Photonic-Integrated-Circuit (PIC) reservoir. Identification of 32 GBd single-polarization signals of 4QAM, 16QAM, 32QAM and 64QAM is...

In order to minimize power usage and maximize spectral efficiency in visible light communication (VLC), we use a multi-objective optimization algorithm and compare DC-biased optical OFDM (DCO-OFDM) with constant envelope OFDM (CE-OFDM)...

Deep-Unfolded Adaptive Projected Subgradient Method for MIMO Detection This paper proposes a MIMO detector based on a deep unfolded superiorized adaptive projected subgradient method (APSM). By learning the design parameters of a superiorized...

Due to the rising threat of deepfakes to security and privacy, it is most important to develop robust and reliable detectors. In this paper, we examine the need for high-quality samples in the training datasets of such detectors. Accordingly, we...

Multimodal user interfaces promise natural and intuitive human–machine interactions. However, is the extra effort for the development of a complex multisensor system justified, or can users also be satisfied with only one input modality? This...

While the evaluation of explanations is an important step towards trustworthy models, it needs to be done carefully, and the employed metrics need to be well-understood. Specifically model randomization testing is often overestimated and regarded...

Applying traditional post-hoc attribution methods to segmentation or object detection predictors offers only limited insights, as the obtained feature attribution maps at input level typically resemble the models' predicted segmentation mask or...

State-of-the-art machine learning models often learn spurious correlations embedded in the training data. This poses risks when deploying these models for high-stake decision-making, such as in medical applications like skin cancer detection. To...

Explainable AI (XAI) is a rapidly evolving field that aims to improve transparency and trustworthiness of AI systems to humans. One of the unsolved challenges in XAI is estimating the performance of these explanation methods for neural networks,...

Explainable Artificial Intelligence (XAI) is an emerging research field bringing transparency to highly complex and opaque machine learning (ML) models. This paper offers a comprehensive overview over techniques that apply XAI practically to...

Sydnone methides represent an almost unknown class of mesoionic compounds which possess exocyclic carbon substituents instead of oxygen (sydnones) or nitrogen (sydnone imines) in the 5-position of a 1,2,3-oxadiazolium ring. Unsubstituted...

Recent work in XAI for eye tracking data has evaluated the suitability of feature attribution methods to explain the output of deep neural sequence models for the task of oculomotric biometric identification. In this work, we employ established...

In this work, we experimentally and numerically compare cyclic shifted constant-amplitude zero-autocorrelation (CAZAC) training sequences (TS) with different number of repetitions, sequence lengths and scalings for channel estimation in an...

In this work we investigate cell damage in whole slice images of the epidermis. A common way for pathologists to annotate a score, characterising the degree of damage for these samples, is the ratio between healthy and unhealthy nuclei. The...