Publications Search Results

Now showing 1 - 10 of 29
  • Publication
    An optimization approach for a milling dynamics simulation based on Quantum Computing
    ( 2024-02-01) ;
    Danz, Sven
    ;
    Kienast, Pascal
    ;
    König, Valentina
    ;
    ;
    Since the machining of complex aerospace components, like integral compressor-rotors (blade integrated disks), is very cost-intensive, optimizing the process by means of process simulations is an active field of research. With the rise of Quantum Computing, a new instrument with high optimization potential is moving into focus. In this paper, a possible application of Quantum Computing for the machining simulation of multi-axis milling of thin-walled aerospace components is discussed. For this reason, a simulation framework used for the milling simulation is analyzed and each component is evaluated separately in relation to Quantum Computing. Parts of the Harrow, Hassidim, and Lloyd algorithm are proposed to enhance the Finite-Element simulation-based component, like the modal analysis for dynamics simulation. This algorithm can solve linear system problems with exponential speed-up over the classical method. The paper presents a roadmap on how the classical steps of a modal analysis for dynamics simulation could be replaced by quantum algorithms based on quantum phase estimation. The implementation of the first working steps is presented to validate this approach. The linear system problem, arising from the dynamics simulation, is analyzed in detail and a minimal value problem of linear coupled oscillators is derived.
  • Publication
    Synthetic Pose Dataset
    ( 2023)
    Sterz, Hannah
    ;
    ;
    Bohné, Thomas
    This thesis explores approaches to synthetic human pose generation, addressing the critical need for diverse and domain-specific datasets to train and enhance models for pose teaching software. With the ever-increasing reliance on machine learning methods that demand extensive training data, the development of synthetic datasets opens up new avenues for tasks such as personalised feedback or pose classification. I investigate two distinct methods for synthetic data creation within the yoga domain. The first approach, data augmentation, employs predefined rules applied to existing motion capture data. This method enables the precise definition of dataset variations according to specific task requirements. The rules applied during generation not only yield plausible poses but also provide valuable labels for training models. The second approach involves training a variational autoencoder, referred to as VAEGen, which generates new synthetic poses by sampling from a latent space and decoding them into pose representations. This method results in more diverse synthetic poses. Furthermore, the latent space representation offers numerous possibilities, including smooth pose transitions and the potential for synthetic movement generation. The application of synthetic data to train models for providing user feedback reveals promising results, with the system accurately predicting the rules governing the difference between two poses, achieving an accuracy rate of 72%. This capability enables valuable feedback to users, aiding them in aligning their poses more consistently with instructorguided positions.
  • Publication
    Detection and Pose Estimation of Flat, Texture-Less Industry Objects on HoloLens Using Synthetic Training
    ( 2023)
    Pöllabauer, Thomas
    ;
    Rücker, Fabian
    ;
    ;
    Current state-of-the-art 6d pose estimation is too compute intensive to be deployed on edge devices, such as Microsoft HoloLens (2) or Apple iPad, both used for an increasing number of augmented reality applications. The quality of AR is greatly dependent on its capabilities to detect and overlay geometry within the scene. We propose a synthetically trained client-server-based augmented reality application, demonstrating state-of-the-art object pose estimation of metallic and texture-less industry objects on edge devices. Synthetic data enables training without real photographs, i.e. for yet-to-be-manufactured objects. Our qualitative evaluation on an AR-assisted sorting task, and quantitative evaluation on both renderings, as well as real-world data recorded on HoloLens 2, sheds light on its real-world applicability.
  • Publication
    Optimization-Based Improvement of Face Image Quality Assessment Techniques
    ( 2023)
    Babnik, Žiga
    ;
    ;
    Štruc, Vitomir
    Contemporary face recognition (FR) models achieve near-ideal recognition performance in constrained settings, yet do not fully translate the performance to unconstrained (real-world) scenarios. To help improve the performance and stability of FR systems in such unconstrained settings, face image quality assessment (FIQA) techniques try to infer sample-quality information from the input face images that can aid with the recognition process. While existing FIQA techniques are able to efficiently capture the differences between high and low quality images, they typically cannot fully distinguish between images of similar quality, leading to lower performance in many scenarios. To address this issue, we present in this paper a supervised quality-label optimization approach, aimed at improving the performance of existing FIQA techniques. The developed optimization procedure infuses additional information (computed with a selected FR model) into the initial quality scores generated with a given FIQA technique to produce better estimates of the "actual" image quality. We evaluate the proposed approach in comprehensive experiments with six state-of-the-art FIQA approaches (CR-FIQA, FaceQAN, SER-FIQ, PCNet, MagFace, SER-FIQ) on five commonly used benchmarks (LFW, CFP-FP, CPLFW, CALFW, XQLFW) using three targeted FR models (ArcFace, ElasticFace, CurricularFace) with highly encouraging results.
  • Publication
    On the Quality and Diversity of Synthetic Face Data and its Relation to the Generator Training Data
    In recent years, advances in deep learning techniques and large-scale identity-labeled datasets have enabled facial recognition algorithms to rapidly gain performance. However, due to privacy issues, ethical concerns, and regulations governing the processing, transmission, and storage of biometric samples, several publicly available face image datasets are being withdrawn by their creators. The reason is that these datasets are mostly crawled from the web with the possibility that not all users had properly consented to processing their biometric data. To mitigate this problem, synthetic face images from generative approaches are motivated to substitute the need for authentic face images to train and test face recognition. In this work, we investigate both the relation between synthetic face image data and the generator authentic training data and the relation between the authentic data and the synthetic data in general under two aspects, i.e. the general image quality and face image quality. The first term refers to perceived image quality and the second measures the utility of a face image for automatic face recognition algorithms. To further quantify these relations, we build the analyses under two terms denoted as the dissimilarity in quality values expressing the general difference in quality distributions and the dissimilarity in quality diversity expressing the diversity in the quality values.
  • Publication
    CR-FIQA: Face Image Quality Assessment by Learning Sample Relative Classifiability
    Face image quality assessment (FIQA) estimates the utility of the captured image in achieving reliable and accurate recognition performance. This work proposes a novel FIQA method, CR-FIQA, that estimates the face image quality of a sample by learning to predict its relative classifiability. This classifiability is measured based on the allocation of the training sample feature representation in angular space with respect to its class center and the nearest negative class center. We experimentally illustrate the correlation between the face image quality and the sample relative classifiability. As such property is only observable for the training dataset, we propose to learn this property by probing internal network observations during the training process and utilizing it to predict the quality of unseen samples. Through extensive evaluation experiments on eight benchmarks and four face recognition models, we demonstrate the superiority of our proposed CR-FIQA over state-of-the-art (SOTA) FIQA algorithms.
  • Publication
    Genetic algorithm optimization for parametrization, digital twinning, and now-casting of unknown small- and medium-scale PV systems based only on on-site measured data
    ( 2023)
    Guzman Razo, Dorian Esteban
    ;
    Madsen, Henrik
    ;
    Accurately predicting and balancing energy generation and consumption are crucial for grid operators and asset managers in a market where renewable energy is increasing. To speed up the process, these predictions should ideally be performed based only on on-site measured data and data available within the monitoring platforms, data which are scarce for small- and medium-scale PV systems. In this study, we propose an algorithm that can now-cast the power output of a photovoltaic (PV) system with high accuracy. Additionally, it offers physical information related to the configuration of such a PV system. We adapted a genetic algorithm-based optimization approach to parametrize a digital twin of unknown PV systems, using only on-site measured PV power and irradiance in the plane of array. We compared several training datasets under various sky conditions. A mean deviation of −1.14 W/kWp and a mean absolute percentage deviation of 1.81% were obtained when we analyzed the accuracy of the PV power now-casting for the year 2020 of the 16 unknown PV systems used for this analysis. This level of accuracy is significant for ensuring the efficient now-casting and operation of PV assets.
  • Publication
    CR-FIQA: Face Image Quality Assessment by Learning Sample Relative Classifiability
    Face image quality assessment (FIQA) estimates the utility of the captured image in achieving reliable and accurate recognition performance. This work proposes a novel FIQA method, CR-FIQA, that estimates the face image quality of a sample by learning to predict its relative classifiability. This classifiability is measured based on the allocation of the training sample feature representation in angular space with respect to its class center and the nearest negative class center. We experimentally illustrate the correlation between the face image quality and the sample relative classifiability. As such property is only observable for the training dataset, we propose to learn this property by probing internal network observations during the training process and utilizing it to predict the quality of unseen samples. Through extensive evaluation experiments on eight benchmarks and four face recognition models, we demonstrate the superiority of our proposed CR-FIQA over state-of-the-art (SOTA) FIQA algorithms. https://github.com/fdbtrs/CR-FIQA
  • Publication
    Pixel-Level Face Image Quality Assessment for Explainable Face Recognition
    In this work, we introduce the concept of pixel-level face image quality that determines the utility of single pixels in a face image for recognition. We propose a training-free approach to assess the pixel-level qualities of a face image given an arbitrary face recognition network. To achieve this, a model-specific quality value of the input image is estimated and used to build a sample-specific quality regression model. Based on this model, quality-based gradients are back-propagated and converted into pixel-level quality estimates. In the experiments, we qualitatively and quantitatively investigated the meaningfulness of our proposed pixel-level qualities based on real and artificial disturbances and by comparing the explanation maps on faces incompliant with the ICAO standards. In all scenarios, the results demonstrate that the proposed solution produces meaningful pixel-level qualities enhancing the interpretability of the face image and its quality. The code is publicly available.
  • Publication
    Impact of Motion Estimation Errors on DVB-S Based Passive ISAR Imaging
    ( 2022)
    Santi, Fabrizio
    ;
    ;
    Pastina, Debora
    ;
    This work explores the effects of inaccuracies on the rotation motion estimation on DVB-S based passive ISAR images focused via Back-Projection Algorithm (BPA). BPA can produce images in a target fixed reference system, thus giving rise to the possibility of directly extracting relevant geometrical features from the ISAR products, such as target width and length; moreover, as the support region does not depend on the particular geometry/frequency, it is suitable for comparing/fusing image products achieved on different channels (even distributed and/or operating on different bands). However, as the exploitation of satellite illuminators can complicate the target motion estimation task, the focusing must deal with inaccurate target kinematics information. In this work, after deriving the theoretical PSF, we derive the image pulse response under estimation errors on the yaw, pitch, and roll motions. Moreover, consequently distortions affecting the image are evaluated via closed-form equations. The theoretical findings are confirmed by experimental data analysis using a passive radar receiving system developed by FHR and an Astra satellite as opportunistic illuminator.