arXiv subjects

Thorsten Jungeblut

Publications and source records attributed to Thorsten Jungeblut.

6 recordsLinked to original sources

arXiv2026

No Attention, No Problem: DPU-Aware Attention Approximation in Modern YOLO on FPGA

Edge-based Artificial Intelligence (AI) acceleration has recently improved progress in real-time object detection. Object detection on edge devices requires a balance between accuracy, speed, and power efficiency. This paper proposes a customized Deep Learning Processor Unit (DPU)-aware architecture for attention-based YOLO variants deployed on AMD FPGAs. Specifically, we evaluate and benchmark YOLOv26 and YOLOv11, two modern attention-based YOLO variants, on the Xilinx ZCU104 across both standard and oriented object detection tasks. We replace unsupported activation functions, substitute split operations with 1x1 convolutions, and approximate the spatial attention mechanism in a DPU-compatible way. All models are then trained and evaluated across six benchmark datasets such as COCO, Pascal VOC, KITTI, DOTA, DIOR-R, and an in-house human presence dataset, and benchmarked across all eight DPU configurations (B512 to B4096) in terms of mAP, FPS, latency, power, and resource utilization. Notably, YOLOv26n and YOLOv26n-OBB deliver the highest end-to-end throughput at 34.05 and 29.55 FPS for standard and oriented detection, respectively, with an average of 5% absolute reduction in accuracy due to quantization while achieving up to approximately 3x lower power consumption compared with the state of the art.

cs.AR↗

arXiv2026

PMOF: A Dataset and Benchmark for Passenger Monitoring Using Overhead Fisheye Cameras

Autonomous staff-free public transport requires reliable in-vehicle passenger monitoring. However, perception inside moving vehicles is challenged by confined spaces, variable illumination, motion-induced background variation, occlusion, and limited viewpoints. To mitigate these spatial constraints, ceiling-mounted fisheye cameras provide full-scene coverage from a single viewpoint. Yet existing public overhead fisheye datasets are recorded in static environments and do not capture the domain shift introduced by vehicle motion. To fill this gap, we introduce PMOF, Passenger Monitoring using Overhead Fisheye cameras, the first public dataset of top-view fisheye imagery captured inside a moving vehicle, comprising over 19k manually annotated frames. PMOF provides rotated bounding boxes, tracking identifiers, and action labels, supporting object detection, tracking, and action recognition. We benchmark PMOF using YOLO26m-obb models fine-tuned under multiple dataset configurations that combine PMOF with existing overhead fisheye datasets. Cross-domain fine-tuning with custom rotation-aware augmentation achieves 94.8% AP50 on PMOF and 96.5% AP50 on an unseen overhead fisheye dataset from a different domain. Our results highlight the domain gap between static and moving environments and show that incorporating PMOF improves detection performance and advances generalization beyond passenger monitoring to broader fisheye-based person detection tasks. The dataset and code are available at https://swermuth.github.io/pmof/.

cs.CV↗

arXiv2025

MALRIS: Malicious Hardware in RIS-Assisted Wireless Communications

Reconfigurable intelligent surfaces (RIS) enhance wireless communication by dynamically shaping the propagation environment, but their integration introduces hardware-level security risks. This paper presents the concept of Malicious RIS (MALRIS), where compromised components behave adversarially, even under passive operation. The focus of this work is on practical threats such as manufacturing time tampering, malicious firmware, and partial element control. Two representative attacks, power-splitting and element-splitting, are modeled to assess their impact. Simulations in a RIS-assisted system reveal that even a limited hardware compromise can significantly degrade performance metrics such as bit error rate, throughput, and secrecy metrics. By exposing this overlooked threat surface, this work aims to promote awareness and support secure, trustworthy RIS deployment in future wireless networks.

eess.SP↗

arXiv2024

A Hybrid Spiking-Convolutional Neural Network Approach for Advancing Machine Learning Models

In this article, we propose a novel standalone hybrid Spiking-Convolutional Neural Network (SC-NN) model and test on using image inpainting tasks. Our approach uses the unique capabilities of SNNs, such as event-based computation and temporal processing, along with the strong representation learning abilities of CNNs, to generate high-quality inpainted images. The model is trained on a custom dataset specifically designed for image inpainting, where missing regions are created using masks. The hybrid model consists of SNNConv2d layers and traditional CNN layers. The SNNConv2d layers implement the leaky integrate-and-fire (LIF) neuron model, capturing spiking behavior, while the CNN layers capture spatial features. In this study, a mean squared error (MSE) loss function demonstrates the training process, where a training loss value of 0.015, indicates accurate performance on the training set and the model achieved a validation loss value as low as 0.0017 on the testing set. Furthermore, extensive experimental results demonstrate state-of-the-art performance, showcasing the potential of integrating temporal dynamics and feature extraction in a single network for image inpainting.

cs.NE↗

arXiv2024

Design-Space Exploration of SNN Models using Application-Specific Multi-Core Architectures

With the motivation and the difficulties that currently exist in comprehending and utilizing the promising features of SNNs, we proposed a novel run-time multi-core architecture-based simulator called "RAVSim" (Runtime Analysis and Visualization Simulator), a cutting-edge SNN simulator, developed using LabVIEW and it is publicly available on their website as an official module. RAVSim is a runtime virtual simulation environment tool that enables the user to interact with the model, observe its behavior of output concentration, and modify the set of parametric values at any time while the simulation is in execution. Recently some popular tools have been presented, but we believe that none of the tools allow users to interact with the model simulation in run time.

cs.NE↗

arXiv2018

Development of Energy Models for Design Space Exploration of Embedded Many-Core Systems

This paper introduces a methodology to develop energy models for the design space exploration of embedded many-core systems. The design process of such systems can benefit from sophisticated models. Software and hardware can be specifically optimized based on comprehensive knowledge about application scenario and hardware behavior. The contribution of our work is an automated framework to estimate the energy consumption at an arbitrary abstraction level without the need to provide further information about the system. We validated our framework with the configurable many-core system CoreVA-MPSoC. Compared to a simulation of the CoreVA-MPSoC on gate level in a 28nm FD-SOI standard cell technology, our framework shows an average estimation error of about 4%.

cs.DC↗