Refine
Year of publication
Document Type
- Conference Proceeding (1253) (remove)
Conference Type
- Konferenzartikel (950)
- Konferenz-Abstract (156)
- Konferenzband (77)
- Sonstiges (42)
- Konferenz-Poster (32)
Language
- English (934)
- German (317)
- Multiple languages (1)
- Russian (1)
Keywords
- Mikroelektronik (62)
- RoboCup (32)
- Gamification (12)
- Machine Learning (12)
- injury (10)
- Biomechanik (9)
- Finite-Elemente-Methode (9)
- Kommunikation (9)
- Assistive Technology (8)
- Produktion (8)
Institute
- Fakultät Elektrotechnik und Informationstechnik (E+I) (bis 03/2019) (453)
- Fakultät Maschinenbau und Verfahrenstechnik (M+V) (286)
- Fakultät Elektrotechnik, Medizintechnik und Informatik (EMI) (ab 04/2019) (213)
- Fakultät Wirtschaft (W) (164)
- Fakultät Medien und Informationswesen (M+I) (bis 21.04.2021) (120)
- ivESK - Institut für verlässliche Embedded Systems und Kommunikationselektronik (113)
- INES - Institut für nachhaltige Energiesysteme (59)
- IMLA - Institute for Machine Learning and Analytics (46)
- ACI - Affective and Cognitive Institute (40)
- Fakultät Medien (M) (ab 22.04.2021) (33)
Open Access
- Open Access (560)
- Closed Access (456)
- Closed (223)
- Bronze (214)
- Diamond (29)
- Grün (13)
- Gold (6)
- Hybrid (6)
The aim of this work is the application and evaluation of a method to visually detect markers at a distance of up to five meters and determine their real-world position. Combinations of cameras and lenses with different parameters were studied to determine the optimal configuration. Based on this configuration, camera images were taken after proper calibration. These images are then transformed into a bird's eye view using a homography matrix. The homography matrix is calculated with four-point pairs as well as with coordinate transformations. The obtained images show the ground plane un distorted, making it possible to convert a pixel position into a real-world position with a conversion factor. The proposed approach helps to effectively create data sets for training neural networks for navigation purposes.
For some years now, additive manufacturing (AM) has offered an alternative to conventional manufacturing processes. The strengths of AM are primarily the rapid implementation of ideas into a usable product and the ability to produce geometrically complex shapes. It has also significantly advanced the lightweight design of products made of plastic. So far, the strength of printed components made of polymers is previously very limited.
Recently, new AM processes have become available that allow the embedding of short and also long fibers in polymer matrix. Thus, the manufacturing of components that provide a significant increase in strength becomes possible. In this way, both complex geometries and sophisticated applications can be implemented. This paper therefore investigates how this new technology can be implemented in product development, focusing on sports equipment. An extensive literature research shows that lightweight design plays a decisive role in sports equipment. In addition, the advantages of AM in terms of individualized products and low quantities can be fully exploited.
An example of this approach is the steering system for a seat sled used by paraplegic athletes in the Olympic discipline of Nordic paraskiing. A particular challenge here is the placement and alignment of the long carbon fibers within the polymer matrix and the verification of the strength by means of Finite-Element-Analysis (FEA). In addition, findings from bionics are used to optimize the lightweight design of the steering system. Using this example, it can be shown that the weight of the steering system can be drastically reduced compared to conventional manufacturing. At the same time, a number of parts can be saved through function integration and thus the manufacturing and assembly effort can be reduced significantly.
Today, Additive Manufacturing (AM) is an important part of teaching for the education of future engineers. Therefore, a variety of approaches have been developed in recent years on how to bring the design for additive manufacturing (DfAM) into university teaching. In a detailed literature review, the advantages and disadvantages of the previous approaches are considered and analysed. Based on this, an extended approach is presented in which students analyse and optimize a given product with respect to additive manufacturing. In doing so, the students have to solve challenging tasks in optimization in product development with the help of methodical approaches and practically implement their developed solutions with state-of-the-art additive processes. To work on this task, the students have two different 3D printers at their disposal, which work with different processes and materials. Thus, the students learn to adapt the design to different manufacturing processes and to consider the restrictions of different materials. The assessment of the results from this course is done through feedback and a written survey.
Systematische Erfassung von Einflussfaktoren für das Additive Tooling von Spritzgusswerkzeugen
(2021)
Additive tooling is a quick and cost-effective way of producing injection molded products and high fidelity prototypes using the injection molding process. As part of product development, additive tooling is integrated into a complex process. A lack of design and application knowledge represents a barrier in its use. The present work shows how a Design-Structure-Matrix (DSM) can be used to systematically record and analyze influencing factors and their interrelationships. A systematic literature search is carried out to identify the factors and relationships.
As a reaction to the increasing market dynamics and complex requirements, today’s products need to be developed quickly and customized to the customer’s individual needs. In the past, CAD systems are mainly used to visualize the model that the product designer creates. Generative Design shifts the task of the CAD program by actively participating in the shaping process. This results in more design options and the complexity of the shapes and geometries increases significantly. This potential can be optimally exploited through the combination of Generative Design with Additive Manufacturing (AM). Artificial intelligence and the input of target parameters generate geometries, for example, by creating material for stressed areas, which in turn develops biomorphic shapes and thus significantly reduces the consumption of resources. This contribution aims at the evaluation of existing applications in CAD systems for generative design. Special attention is paid to the requirements in design education and easy access for students. For this purpose, three representative CAD systems are selected and analyzed with the help of a comprehensive example of mass reduction. The aim is to perform an individual result analysis in order to assess the application based on various criteria. By using different materials, the influence of the material for the generation is investigated by comparing the material distribution. By comparing the generated models, differences of the CAD systems can be identified and possible fields of application can be presented. By specifying the manufacturing parameters for the generation of the models, the feasibility of AM can be guaranteed without having to modify the results. The physical implementation of the example by means of Fused Deposition Modeling demonstrates this in an exemplary way and examines the interface of the Generative Design and AM. The results of this contribution will enable an evaluation of the different CAD systems for Generative Design according to technical, visual and economic aspects.
This paper presents the development of an energy harvesting solution for a driven tool holder. The tool holder environment was analysed, a test stand built and the designed electromagnetic rotation harvester was evaluated. The reported harvester is based on low cost off-the-shelf components and 3D printed parts. The utilisation of SMD coils allows easy adaptation to changing parameters of the integration area. Energy harvesting in tool holders enables predictive maintenance or condition monitoring in the industrial production. These capabilities are mandatory nowadays in regards of IIoT. A reliable energy source is key for continuous monitoring. Changing batteries becomes obsolete. The results provide useful insight for future harvesters.
Engineering, construction and operation of complex machines involves a wide range of complicated, simultaneous tasks, which potentially could be automated. In this work, we focus on perception tasks in such systems, investigating deep learning approaches for multi-task transfer learning with limited training data. We show an approach that takes advantage of a technical systems’ focus on selected objects and their properties. We create focused representations and simultaneously solve joint objectives in a system through multi-task learning with convolutional autoencoders. The focused representations are used as a starting point for the data-saving solution of the additional tasks. The efficiency of this approach is demonstrated using images and tasks of an autonomous circular crane with a grapple.
An Empirical Investigation of Model-to-Model Distribution Shifts in Trained Convolutional Filters
(2021)
We present first empirical results from our ongoing investigation of distribution shifts in image data used for various computer vision tasks. Instead of analyzing the original training and test data, we propose to study shifts in the learned weights of trained models. In this work, we focus on the properties of the distributions of dominantly used 3x3 convolution filter kernels. We collected and publicly provide a data set with over half a billion filters from hundreds of trained CNNs, using a wide range of data sets, architectures, and vision tasks. Our analysis shows interesting distribution shifts (or the lack thereof) between trained filters along different axes of meta-parameters, like data type, task, architecture, or layer depth. We argue, that the observed properties are a valuable source for further investigation into a better understanding of the impact of shifts in the input data to the generalization abilities of CNN models and novel methods for more robust transfer-learning in this domain.
A fundamental and still largely unsolved question in the context of Generative Adversarial Networks is whether they are truly able to capture the real data distribution and, consequently, to sample from it. In particular, the multidimensional nature of image distributions leads to a complex evaluation of the diversity of GAN distributions. Existing approaches provide only a partial understanding of this issue, leaving the question unanswered. In this work, we introduce a loop-training scheme for the systematic investigation of observable shifts between the distributions of real training data and GAN generated data. Additionally, we introduce several bounded measures for distribution shifts, which are both easy to compute and to interpret. Overall, the combination of these methods allows an explorative investigation of innate limitations of current GAN algorithms. Our experiments on different data-sets and multiple state-of-the-art GAN architectures show large shifts between input and output distributions, showing that existing theoretical guarantees towards the convergence of output distributions appear not to be holding in practice.
Correlation Clustering, also called the minimum cost Multicut problem, is the process of grouping data by pairwise similarities. It has proven to be effective on clustering problems, where the number of classes is unknown. However, not only is the Multicut problem NP-hard, an undirected graph G with n vertices representing single images has at most edges, thus making it challenging to implement correlation clustering for large datasets. In this work, we propose Multi-Stage Multicuts (MSM) as a scalable approach for image clustering. Specifically, we solve minimum cost Multicut problems across multiple distributed compute units. Our approach not only allows to solve problem instances which are too large to fit into the shared memory of a single compute node, but it also achieves significant speedups while preserving the clustering accuracy at the same time. We evaluate our proposed method on the CIFAR10 …
Aerosol particles play an important role in the climate system by absorbing and scattering radiation and influencing cloud properties. They are also one of the biggest sources of uncertainty for climate modeling. Many climate models do not include aerosols in sufficient detail. In order to achieve higher accuracy, aerosol microphysical properties and processes have to be accounted for. This is done in the ECHAM-HAM global climate aerosol model using the M7 microphysics model, but increased computational costs make it very expensive to run at higher resolutions or for a longer time. We aim to use machine learning to approximate the microphysics model at sufficient accuracy and reduce the computational cost by being fast at inference time. The original M7 model is used to generate data of input-output pairs to train a neural network on it. By using a special logarithmic transform we are able to learn the variables tendencies achieving an average score of . On a GPU we achieve a speed-up of 120 compared to the original model.
Recently, adversarial attacks on image classification networks by the AutoAttack (Croce and Hein, 2020b) framework have drawn a lot of attention. While AutoAttack has shown a very high attack success rate, most defense approaches are focusing on network hardening and robustness enhancements, like adversarial training. This way, the currently best-reported method can withstand about 66% of adversarial examples on CIFAR10. In this paper, we investigate the spatial and frequency domain properties of AutoAttack and propose an alternative defense. Instead of hardening a network, we detect adversarial attacks during inference, rejecting manipulated inputs. Based on a rather simple and fast analysis in the frequency domain, we introduce two different detection algorithms. First, a black box detector that only operates on the input images and achieves a detection accuracy of 100% on the AutoAttack CIFAR10 benchmark and 99.3% on ImageNet, for epsilon = 8/255 in both cases. Second, a whitebox detector using an analysis of CNN feature maps, leading to a detection rate of also 100% and 98.7% on the same benchmarks.
Transformer models have recently attracted much interest from computer vision researchers and have since been successfully employed for several problems traditionally addressed with convolutional neural networks. At the same time, image synthesis using generative adversarial networks (GANs) has drastically improved over the last few years. The recently proposed TransGAN is the first GAN using only transformer-based architectures and achieves competitive results when compared to convolutional GANs. However, since transformers are data-hungry architectures, TransGAN requires data augmentation, an auxiliary super-resolution task during training, and a masking prior to guide the self-attention mechanism. In this paper, we study the combination of a transformer-based generator and convolutional discriminator and successfully remove the need of the aforementioned required design choices. We evaluate our approach by conducting a benchmark of well-known CNN discriminators, ablate the size of the transformer-based generator, and show that combining both architectural elements into a hybrid model leads to better results. Furthermore, we investigate the frequency spectrum properties of generated images and observe that our model retains the benefits of an attention based generator.
Generative adversarial networks are the state of the art approach towards learned synthetic image generation. Although early successes were mostly unsupervised, bit by bit, this trend has been superseded by approaches based on labelled data. These supervised methods allow a much finer-grained control of the output image, offering more flexibility and stability. Nevertheless, the main drawback of such models is the necessity of annotated data. In this work, we introduce an novel framework that benefits from two popular learning techniques, adversarial training and representation learning, and takes a step towards unsupervised conditional GANs. In particular, our approach exploits the structure of a latent space (learned by the representation learning) and employs it to condition the generative model. In this way, we break the traditional dependency between condition and label, substituting the latter by unsupervised features coming from the latent space. Finally, we show that this new technique is able to produce samples on demand keeping the quality of its supervised counterpart.
Generative adversarial networks (GANs) provide state-of-the-art results in image generation. However, despite being so powerful, they still remain very challenging to train. This is in particular caused by their highly non-convex optimization space leading to a number of instabilities. Among them, mode collapse stands out as one of the most daunting ones. This undesirable event occurs when the model can only fit a few modes of the data distribution, while ignoring the majority of them. In this work, we combat mode collapse using second-order gradient information. To do so, we analyse the loss surface through its Hessian eigenvalues, and show that mode collapse is related to the convergence towards sharp minima. In particular, we observe how the eigenvalues of the are directly correlated with the occurrence of mode collapse. Finally, motivated by these findings, we design a new optimization algorithm called nudged-Adam (NuGAN) that uses spectral information to overcome mode collapse, leading to empirically more stable convergence properties.
In this preliminary report, we present a simple but very effective technique to stabilize the training of CNN based GANs. Motivated by recently published methods using frequency decomposition of convolutions (eg Octave Convolutions), we propose a novel convolution scheme to stabilize the training and reduce the likelihood of a mode collapse. The basic idea of our approach is to split convolutional filters into additive high and low frequency parts, while shifting weight updates from low to high during the training. Intuitively, this method forces GANs to learn low frequency coarse image structures before descending into fine (high frequency) details. Our approach is orthogonal and complementary to existing stabilization methods and can simply plugged into any CNN based GAN architecture. First experiments on the CelebA dataset show the effectiveness of the proposed method.
We demonstrate how to exploit group sparsity in order to bridge the areas of network pruning and neural architecture search (NAS). This results in a new one-shot NAS optimizer that casts the problem as a single-level optimization problem and does not suffer any performance degradation from discretizating the architecture.
Despite the success of convolutional neural networks (CNNs) in many computer vision and image analysis tasks, they remain vulnerable against so-called adversarial attacks: Small, crafted perturbations in the input images can lead to false predictions. A possible defense is to detect adversarial examples. In this work, we show how analysis in the Fourier domain of input images and feature maps can be used to distinguish benign test samples from adversarial images. We propose two novel detection methods: Our first method employs the magnitude spectrum of the input images to detect an adversarial attack. This simple and robust classifier can successfully detect adversarial perturbations of three commonly used attack methods. The second method builds upon the first and additionally extracts the phase of Fourier coefficients of feature-maps at different layers of the network. With this extension, we are able to improve adversarial detection rates compared to state-of-the-art detectors on five different attack methods. The code for the methods proposed in the paper is available at github.com/paulaharder/SpectralAdversarialDefense
In this work, we evaluate two different image clustering objectives, k-means clustering and correlation clustering, in the context of Triplet Loss induced feature space embeddings. Specifically, we train a convolutional neural network to learn discriminative features by optimizing two popular versions of the Triplet Loss in order to study their clustering properties under the assumption of noisy labels. Additionally, we propose a new, simple Triplet Loss formulation, which shows desirable properties with respect to formal clustering objectives and outperforms the existing methods. We evaluate all three Triplet loss formulations for K-means and correlation clustering on the CIFAR-10 image classification dataset.
The term “attribute transfer” refers to the tasks of altering images in such a way, that the semantic interpretation of a given input image is shifted towards an intended direction, which is quantified by semantic attributes. Prominent example applications are photo realistic changes of facial features and expressions, like changing the hair color, adding a smile, enlarging the nose or altering the entire context of a scene, like transforming a summer landscape into a winter panorama. Recent advances in attribute transfer are mostly based on generative deep neural networks, using various techniques to manipulate images in the latent space of the generator. In this paper, we present a novel method for the common sub-task of local attribute transfers, where only parts of a face have to be altered in order to achieve semantic changes (e.g. removing a mustache). In contrast to previous methods, where such local changes have been implemented by generating new (global) images, we propose to formulate local attribute transfers as an inpainting problem. Removing and regenerating only parts of images, our “Attribute Transfer Inpainting Generative Adversarial Network” (ATI-GAN) is able to utilize local context information to focus on the attributes while keeping the background unmodified resulting in visually sound results.