Recent publications

We present a novel method ensuring the right reasons on the concept level by reducing the model's sensitivity towards biases through the gradient. When modeling biases via Concept Activation Vectors, we highlight the importance of choosing robust...

Solid-state optical beam-steering utilizing polymer waveguides as edge emitters to form optical phased arrays (OPAs) with aperiodic spacing for operation at 1550 nm is demonstrated for the first time. Power consumption of 1.28 mW/? per channel is...

We introduce a novel NeRF-based framework for pose-dependent rendering of human performances where the radiance field is warped around an SMPL body mesh, thereby creating a new surface-aligned representation. Our representation can be animated...

This study investigates the effectiveness of neural network architectures in hyperspectral image demosaicing. We introduce a range of network models and modifications, and compare them with classical interpolation methods and existing reference...

Our method builds on existing state-of-the-art 3D GAN inversion techniques to allow for consistent and simultaneous inversion of multiple views of the same subject. We employ a multi-latent extension to handle inconsistencies present in dynamic...

We propose a generative deep learning network for texture map super-resolution using a differentiable renderer and calibrated reference images. Combining a super-resolution generative adversarial network (GAN) with differentiable rendering, we...

We propose a method for automatic prevention of ghosting artifacts based on a pixel-wise alignment during morph generation. We evaluate our proposed method on state-of-the-art detectors and show that our morphs are harder to detect, particularly,...

We present a novel approach combining two calibrated multispectral real-time capable snapshot cameras, covering different spectral ranges, into a stereo-system. Therefore, a hyperspectral data-cube can be continuously captured. The combined use...

We validate experimentally the claim that, in sub-THz mobile access networks, single-carrier or low-number-of-subcarriers modulations are very attractive competitors to the dramatically more complex and energy-inefficient traditional...

This work integrates an evanescent Bragg grating sensor into a polymer waveguide with microchannels. The sensor, built with epoxide-based polymers, is characterized through chemical applications. Temperature sensitivity is demonstrated (-47.75...

We explored how explainable artificial intelligence (XAI) can help to shed light into the inner workings of neural networks for protein function prediction, by extending the widely used XAI method of integrated gradients such that latent...

Recent attention to semantic communication, driven by task-oriented solutions, aims to optimize resource use. Despite perceived efficiency gains, few practical implementations exist. This paper revisits theoretical foundations, emphasizing...

We present iV2V and iV2i+, two machine-learning datasets for industrial wireless communication. The datasets cover sidelink and cellular communication involving autonomous robots together with localization and sensing data, which can be used to...

We demonstrate a compact heterodyne Laser Doppler Vibrometer (LDV) based on the realization of optical frequency shift in the silicon nitride photonic integration ! platform (TriPleX). The system comprises a dual-polarization coherent detector...

This paper presents a hybrid photonic integration concept based on the use of a polymer motherboard, InP EML arrays and InP PD arrays to realize WDM and SDM Terabit optical engines operating at 100-Gb/s or even at 20! 0-Gb/s per lane. The optical...

Bi-substituted acetylenes with a quinolinium and an isoquinolinium substituent are described, which reversibly form intensely colored adducts with O-nucleophiles and thus enable the detection of >0,5 ppm hydroxide on the surfaces of various...

Explainable Artificial Intelligence (XAI) is targeted at understanding how models perform feature selection and derive their classification decisions. This paper explores post-hoc explanations for deep neural networks in the audio domain....

Experimental tests of the OFDM and OOK-based PHYs in the Li-Fi standard IEEE 802.15.13-2023 show that the higher peak data rates of OFDM and the longer reach of OOK make them suitable for deployment in down- and uplink use cases, respectively.

This paper addresses the self-interference problem in full duplex operation by leveraging an intelligent reflecting surface (IRS) to provide spatial separation of the transmitted and the received (i.e. reflected) signal in a monostatic sensing...

We demonstrate data rates up to 4 Gbit/s with a VCSEL-based transmitter prototype using pulsed on-off-keying modulation for high-speed LiFi with a large coverage area.

In recent years, immersive applications such as Cloud XR have emerged, which require very low latency to ensure a high quality of experience. This paper presents a congestion control algorithm combined with L4S to achieve stable and low latency...

We present a guard-ring free InGaAs/InP single photon avalanche diode with 20 µm diameter for the optical C-band. At 225 K, 25.6 µs dead time and 17% detection efficiency, the dark count rate is 3 kcps with 0.5% afterpulsing probability. This...

Due to the lack of a feedback channel in the C-V2X sidelink, finding a suitable MCS level is a difficult task. In this paper, we propose an ML approach that uses quantile prediction to predict the MCS level with the highest achievable data rate....

This paper presents a hybrid photonic integration concept based on the use of a polymer motherboard, InP EML arrays and InP PD arrays to realize WDM and SDM Terabit optical engines operating at 100-Gb/s or even at 20! 0-Gb/s per lane. The optical...

One of cost-effective ways to increase the transmission capacity of current standard wavelength division multiplexing (WDM) transmission systems is to use a wavelength band other than the C-band to transmit in multi-band. We proposed the concept...

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...

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 introduce an XAI...

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...

Due to their perfect periodic autocorrelation property, Zadoff–Chu sequences are often used as stimulus signals in the measurement of radio channel responses. In this letter, the cross-correlation of a linear shift-invariant system's response to...