AImotion Bavaria
Refine
Document Type
- Conference Paper (86)
- Article (80)
- Preprint (25)
- Book (3)
- Part of a Compilation / Book Chapter (3)
- Report (2)
Institute
- AImotion Bavaria (199)
- Fakultät Informatik (110)
- Fakultät Elektro- und Informationstechnik (41)
- Fakultät Wirtschaftsingenieurwesen (36)
- Fakultät Maschinenbau (6)
- CARISSMA Institute of Automated Driving (C-IAD) (5)
- Fraunhofer-Anwendungszentrum "Vernetzte Mobilität und Infrastruktur" (4)
- Institut für Innovative Mobilität (IIMo) (4)
- THI Business School (3)
- CARISSMA Institute of Electric, Connected and Secure Mobility (C-ECOS) (2)
Review
- peer-review (149)
- nein (33)
- editorial review (7)
- ja (2)
Version
- published (81)
The QUILT-1M dataset is the first openly available dataset containing images harvested from various online sources. While it provides a huge data variety, the image quality and composition is highly heterogeneous, impacting its utility for text-conditional image synthesis. We propose an automatic pipeline that provides predictions of the most common impurities within the images, e.g., visibility of narrators, desktop environment and pathology software, or text within the image. Additionally, we propose to use semantic alignment filtering of the image-text pairs. Our findings demonstrate that by rigorously filtering the dataset, there is a substantial enhancement of image fidelity in text-to-image tasks.
For the use of the CSS model, workflows are needed to describe, assign and derive capabilities. This contribution provides a Capability Determining method that encompasses the three underlying technology-independent methods to guide stakeholders in describing, assigning and deriving capabilities for manufacturing. The first underlying method, Describing Capabilities, describes a workflow for the formal semantic description of capabilities. The workflow for assigning capabilities to a production resource is explained using the second method Assigning Capability. In the last workflow, capabilities are derived from production specification or a process description using the method Deriving Capability.
Enhancing traffic signal optimisation has the potential to improve urban traffic flow without the need for expensive infrastructure modifications. While reinforcement learning (RL) techniques have demonstrated their effectiveness in simulations, their real-world implementation is still a challenge. Real-world systems need to be developed that guarantee a deployable action definition for real traffic systems while prioritising safety constraints and robust policies. This paper introduces a method to overcome this challenge by introducing a novel action definition that optimises parameter-level control programmes designed by traffic engineers. The complete proposed framework consists of a traffic situation estimation, a feature extractor, and a system that enables training on estimates of real-world traffic situations. Further multimodal optimisation, scalability, and continuous training after deployment could be achieved. The first simulative tests using this action definition show an average improvement of more than 20% in traffic flow compared to the baseline – the corresponding pre-optimised real-world control.
NoiSLU: a Noisy speech corpus for Spoken Language Understanding in the Public Transport Domain
(2024)
Hierarchical Multi-task Learning with Articulatory Attributes for Cross-Lingual Phoneme Recognition
(2023)
Multidimensional Magnetic Resonance Imaging (MRI) is a versatile tool for microstructure mapping. We use a diffusion weighted inversion-recovery spin echo (DW-IR-SE) sequence with spiral readouts at ultra-strong gradients to acquire a rich diffusion-relaxation data set with sensitivity to myelin water. We reconstruct 1D and 2D spectra with a two-step convex optimization approach and investigate a variety of multidimensional MRI methods, including 1D multi-component relaxometry, 1D multi-component diffusometry, 2D relaxation correlation imaging, and 2D diffusion-relaxation correlation spectroscopic imaging (DR-CSI), in terms of their potential to quantify tissue microstructure, including the myelin water fraction (MWF). We observe a distinct spectral peak that we attribute to myelin water in multi-component T1 relaxometry, T1-T2 correlation, T1-D correlation, and T2-D correlation imaging. Due to lower achievable echo times compared to diffusometry, MWF maps from relaxometry have higher quality. While 1D multi-component T1 data allows much faster myelin mapping, 2D approaches could offer unique insights into tissue microstructure and especially myelin diffusion.
Removing the bias and variance of multicentre data has always been a challenge in large scale digital healthcare studies, which requires the ability to integrate clinical features extracted from data acquired by different scanners and protocols to improve stability and robustness. Previous studies have described various computational approaches to fuse single modality multicentre datasets. However, these surveys rarely focused on evaluation metrics and lacked a checklist for computational data harmonisation studies. In this systematic review, we summarise the computational data harmonisation approaches for multi-modality data in the digital healthcare field, including harmonisation strategies and evaluation metrics based on different theories. In addition, a comprehensive checklist that summarises common practices for data harmonisation studies is proposed to guide researchers to report their research findings more effectively. Last but not least, flowcharts presenting possible ways for methodology and metric selection are proposed and the limitations of different methods have been surveyed for future research.
With the rise of autonomous driving technology, concerns over the safety of AI-based systems have increased, particularly regarding their effect on human life and infrastructure. The complexity and unpredictable nature of AI-enabled software functions have disrupted traditional safety evaluation methods. In such a scenario, the quality control methods used at the manufacturing environments of sensors used for deploying AI-based perception functions need to be coupled with the downstream functions. This study focuses on investigating the impact of a specific production-related defect called blemish, which is caused by debris on image sensors of camera modules. The research aims to answer two fundamental questions: 1) How can blemish artifacts be mathematically modeled, and what is the process for developing a suitable dataset that incorporates this defect? 2) How can the influence of blemish artifacts on performance of object detection models be traced back to define quality requirements of the camera modules? This work takes a novel perspective by shifting the focus from improving model resilience to improving hardware quality camera modules by studying the effects of manufacturing defects on downstream functions. This research seeks to establish a framework to derive quality targets for front camera module manufacturing which can be extended to other sensors for similar applications.
Accurate and comprehensive semantic segmentation of Bird's Eye View (BEV) is essential for ensuring safe and proactive navigation in autonomous driving. Although cooperative perception has exceeded the detection capabilities of single-agent systems, prevalent camera-based algorithms in cooperative perception neglect valuable information derived from historical observations. This limitation becomes critical during sensor failures or communication issues as cooperative perception reverts to single-agent perception, leading to degraded performance and incomplete BEV segmentation maps. This paper introduces TempCoBEV, a temporal module designed to incorporate historical cues into current observations, thereby improving the quality and reliability of BEV map segmentations. We propose an importance-guided attention architecture to effectively integrate temporal information that prioritizes relevant properties for BEV map segmentation. TempCoBEV is an independent temporal module that seamlessly integrates into state-of-the-art camera-based cooperative perception models. We demonstrate through extensive experiments on the OPV2V dataset that TempCoBEV performs better than non-temporal models in predicting current and future BEV map segmentations, particularly in scenarios involving communication failures. We show the efficacy of TempCoBEV and its capability to integrate historical cues into the current BEV map, improving predictions under optimal communication conditions by up to 2% and under communication failures by up to 19%. The code will be published on GitHub.
Compression of end-to-end non-autoregressive image-to-speech system for low-resourced devices
(2023)
Numerous prognostic factors are currently assessed histopathologically in biopsies of canine mast cell tumors to evaluate clinical behavior. In addition, PCR analysis of the c-Kit exon 11 mutational status is often performed to evaluate the potential success of a tyrosine kinase inhibitor therapy. This project aimed at training deep learning models (DLMs) to identify the c-Kit-11 mutational status of MCTs solely based on morphology without additional molecular analysis. HE slides of 195 mutated and 173 non-mutated tumors were stained consecutively in two different laboratories and scanned with three different slide scanners. This resulted in six different datasets (stain-scanner variations) of whole slide images. DLMs were trained with single and mixed datasets and their performances was assessed under scanner and staining domain shifts. The DLMs correctly classified HE slides according to their c-Kit 11 mutation status in, on average, 87% of cases for the best-suited stain-scanner variant. A relevant performance drop could be observed when the stain-scanner combination of the training and test dataset differed. Multi-variant datasets improved the average accuracy but did not reach the maximum accuracy of algorithms trained and tested on the same stain-scanner variant. In summary, DLM-assisted morphological examination of MCTs can predict c-Kit-exon 11 mutational status of MCTs with high accuracy. However, the recognition performance is impeded by a change of scanner or staining protocol. Larger data sets with higher numbers of scans originating from different laboratories and scanners may lead to more robust DLMs to identify c-Kit mutations in HE slides.
Emergent Cooperation from Mutual Acknowledgment Exchange in Multi-Agent Reinforcement Learning
(2022)
Peer incentivization (PI) is a recent approach, where all agents learn to reward or to penalize each other in a distributed fashion which often leads to emergent cooperation. Current PI mechanisms implicitly assume a flawless communication channel in order to exchange rewards. These rewards are directly integrated into the learning process without any chance to respond with feedback. Furthermore, most PI approaches rely on global information which limits scalability and applicability to real-world scenarios, where only local information is accessible. In this paper, we propose Mutual Acknowledgment Token Exchange (MATE), a PI approach defined by a two-phase communication protocol to mutually exchange acknowledgment tokens to shape individual rewards. Each agent evaluates the monotonic improvement of its individual situation in order to accept or reject acknowledgment requests from other agents. MATE is completely decentralized and only requires local communication and information. We evaluate MATE in three social dilemma domains. Our results show that MATE is able to achieve and maintain significantly higher levels of cooperation than previous PI approaches. In addition, we evaluate the robustness of MATE in more realistic scenarios, where agents can defect from the protocol and where communication failures can occur. We also evaluate the sensitivity of MATE w.r.t. the choice of token values.
Diagnosis of malignancy in oropharyngeal confocal laser endomicroscopy using GPT 4.0 with vision
(2024)
AbstractGraph data models enable efficient storage, visualization, and analysis of highly interlinked data, by providing the benefits of horizontal scalability and high query performance. Clustering techniques, such as K-means, hierarchical clustering, are highly beneficial tools in data mining and machine learning to find meaningful similarities and differences between data points. Recent developments in graph data models, as well as clustering algorithms for graph data, have shown promising results in image segmentation, gene data analysis, etc. This has been primarily achieved through research and development of algorithms in the field of spectral theory, leading to the conception of spectral clustering algorithms. Spectral clustering algorithms have been one of the most effective in grouping similar data points in graph data models. In this paper, we have compiled 16 spectral clustering algorithms and compared their computational complexities, after an overview of graph data models and graph database models. Furthermore, we provided a broad taxonomy to classify most existing clustering algorithms and discussed the taxonomy in detail.
Knowledge Tracing (KT) aims to predict the future performance of students by tracking the development of their knowledge states. Despite all the recent progress made in this field, the application of KT models in education systems is still restricted from the data perspectives: 1) limited access to real life data due to data protection concerns, 2) lack of diversity in public datasets, 3) noises in benchmark datasets such as duplicate records. To resolve these problems, we simulated student data with three statistical strategies based on public datasets and tested their performance on two KT baselines. While we observe only minor performance improvement with additional synthetic data, our work shows that using only synthetic data for training can lead to similar performance as real data.
ZusammenfassungIm BMBF-Verbundprojekt HAnS entwickeln und implementieren neun Hochschulen sowie drei hochschulübergreifende Einrichtungen ein intelligentes Hochschul-Assistenz-System als Open-Source-Lösung. Videobasierte Lehrmaterialien werden verschriftlicht und durch eine Indexierung Stichwortsuchen ermöglicht; geplant ist, über einen KI-Tutor automatisiert Übungsaufgaben zu generieren. Studierende sollen so in ihrem Selbststudium digital unterstützt werden. Die technische Entwicklung wird interdisziplinär – auch sozialwissenschaftlich und pädagogisch – begleitet und in einem iterativen Vorgehen evidenzbasiert entsprechend Design-Based-Research angepasst. Wissen und Wertesystem der Anwender*innen, Didaktik, Ethik, Akzeptanz und Datenschutz werden dabei im Entwicklungsprozess einbezogen.
History has shown that attitudes toward new technologies can change abruptly following disruptive events. During the COVID-19 pandemic, it became apparent that care robots enable increased social isolation. This feature of robotic care usually raises strong ethical concerns about potentially decreased comfort for the care-dependent. In a large-scale online study, we tested the influence of the pandemic on people's affective attitudes toward care robots. In vignettes on different care scenarios, we measured participants' perceived comfort levels in situations with care robots and human caregivers while controlling for their fear of infection with a viral disease. We found that people generally feel less comfortable with a care robot than with a human caregiver. However, those who had a strong fear of being infected during the pandemic did not devalue a care robot compared to a human caregiver. While care robots remain ethically contested, this study shows that affective attitudes toward care robots may change significantly if they can address an urgent need.
Aviation statistics identify collision with terrain and obstacles as a leading cause of helicopter accidents. Assisting helicopter pilots in detecting the presence of obstacles can greatly mitigate the risk of collisions. However, only a limited number of helicopters in operation have an installed helicopter terrain awareness and warning system (HTAWS), while the cost of active obstacle warning systems remains prohibitive for many civil operators. In this work, we apply machine learning to automate obstacle detection and classification in combination with commercially available airborne optical sensors. While numerous techniques for learning-based object detection have been published in the literature, many of them are data and computation intensive. Our approach seeks to balance the detection and classification accuracy of the method with the size of the training data required and the runtime. Specifically, our approach combines the invariant feature extraction ability of pretrained deep convolutional neural networks (CNNs) and the high-speed training and classification ability of a novel, proprietary frequency-domain support vector machine (SVM) method. We describe our experimental setup comprising the CNN + SVM model and datasets of predefined classes of obstacles—pylons, chimneys, antennas, TV towers, wind turbines, helicopters—synthesized from prerecorded airborne video sequences of low-altitude helicopter flight. We analyze the detection performance using average precision, average recall, and runtime performance metrics on representative test data. Finally, we present a simple architecture for real-time, onboard implementation and discuss the obstacle detection performance of recently concluded flight tests.
How would people distribute risks of autonomous vehicles (AVs) in everyday road traffic? The rich literature on the ethics of autonomous vehicles (AVs) revolves around moral judgments in unavoidable collision scenarios. We argue for extending the debate to driving behaviors in everyday road traffic where ubiquitous ethical questions arise due to the permanent redistribution of risk among road users. This distribution of risks raises ethically relevant questions that cannot be evaded by simple heuristics such as “hitting the brakes.” Using an interactive, graphical representation of different traffic situations, we measured participants’ preferences on driving maneuvers of AVs in a representative survey in Germany. Our participants’ preferences deviated significantly from mere collision avoidance. Interestingly, our participants were willing to take risks themselves for the benefit of other road users, suggesting that the social dilemma of AVs may be mitigated in risky environments. Our research might build a bridge between engineers and philosophers to discuss the ethics of AVs more constructively.
In the realm of prognostics and health management (PHM), it is common to possess not only process data but also domain knowledge, which, if integrated into data-driven algorithms, can aid in solving specific tasks. This paper explores the integration of knowledge graphs (KGs) into deep learning models to develop a more resilient approach capable of handling domain shifts, such as variations in machine operating conditions. We present and assess a KG-enhanced deep learning approach in a representative PHM use case, demonstrating its effectiveness by incorporating domain-invariant knowledge through the KG. Furthermore, we provide guidance for constructing a comprehensive hierarchical KG representation that preserves semantic information while facilitating numerical representation. The experimental results showcase the improved performance and domain shift robustness of the KG-enhanced approach in fault diagnostics.
AbstractChatGPT is not only fun to chat with, but it also searches information, answers questions, and gives advice. With consistent moral advice, it can improve the moral judgment and decisions of users. Unfortunately, ChatGPT’s advice is not consistent. Nonetheless, it does influence users’ moral judgment, we find in an experiment, even if they know they are advised by a chatting bot, and they underestimate how much they are influenced. Thus, ChatGPT corrupts rather than improves its users’ moral judgment. While these findings call for better design of ChatGPT and similar bots, we also propose training to improve users’ digital literacy as a remedy. Transparency, however, is not sufficient to enable the responsible use of AI.