52039 research outputs found
Sort by
A Synergy of Computer Graphics and Generative AI: Advancements and Challenges
No full textA traditional computer graphics domain has received an unprecedented boost from the newest developments in
generative Artificial Intelligence (GenAI). It affects all areas: from image generation, to face recognition, to
object detection, to aerial surveillance, to autonomous car vision systems. The newest deep learning architectures
make it possible to generate new images from texts, to apply styles to portraits, to de-identify facial images, and
to recognize human and objects in videos. This keynote will delve into some of the most exciting applications in
medical AI diagnostics, human face recognition and aesthetics domains, while making a strong case for resulting
image authenticity, bias mitigation, and trus
Improving Image Reconstruction using Incremental PCA-Embedded Convolutional Variational Auto-Encoder
No full textTraditional image reconstruction methods often face challenges like noise, artifacts, and blurriness, requiring
handcrafted algorithms for effective resolution. In contrast, deep learning techniques, notably Convolutional
Neural Networks (CNNs) and Variational Autoencoders (VAEs), present more robust alternatives. This paper
presents a novel and efficient approach for image reconstruction employing Convolutional Variational Autoen coders (CVAEs). We use Incremental Principal Component Analysis (IPCA) to enhance efficiency by discerning
and capturing significant features within the latent space. This model is integrated into both the encoder and
sampling stages of CVAEs, refining their capability to generate high-fidelity images. Our incremental strategy
mitigates scalability issues associated with traditional PCA while preserving the model’s aptitude for identifying
crucial image features. Experimental validation utilizing the MNIST dataset showcases noteworthy reductions in
processing time and enhancements in image quality, underscoring the efficacy and potential applicability of our
model for large-scale image generation tasks
3D Multi-Criteria Design Generation and Optimization of an Engine Mount for an Unmanned Air Vehicle Using a Conditional Variational Autoencoder
No full textThis research as part of the project LaiLa is funded
by dtec.bw - Digitalization and Technology Reasearch
Center of the Bundeswehr which we gratefully ac knowledge. dtec.bw is funded by the European Union
- NextGenerationEU.One of the most promising developments in computer vision in recent years is the use of generative neural net works for functionality condition-based 3D design reconstruction and generation. Here, neural networks learn
dependencies between functionalities and a geometry in a very effective way. For a neural network the function alities are translated in conditions to a certain geometry. But the more conditions the design generation needs
to reflect, the more difficult it is to learn clear dependencies. This leads to a multi criteria design problem due
various conditions, which are not considered in the neural network structure so far. In this paper, we address this
multi-criteria challenge for a 3D design use case related to an unmanned aerial vehicle (UAV) motor mount. We
generate 10,000 abstract 3D designs and subject them all to simulations for three physical disciplines: mechanics,
thermodynamics, and aerodynamics. Then, we train a Conditional Variational Autoencoder (CVAE) using the
geometry and corresponding multicriteria functional constraints as input. We use our trained CVAE as well as
the Marching cubes algorithm to generate meshes for simulation based evaluation. The results are then evaluated
with the generated UAV designs. Subsequently, we demonstrate the ability to generate optimized designs under
self-defined functionality conditions using the trained neural networ
Impact of Calibration Matrices on 3D Monocular Object Detection: Filtering, Dataset Combination and Integration of Synthetic Data
No full textThis research is funded and supported by SEGULA
Technologies. We would like to thank SEGULA Tech nologies for their collaboration and for allowing us to
conduct this research. We would like to thank also
the engineers of the Autonomous Navigation Laboratory (ANL) of IRSEEM for their support. In addition,
this work was performed, in part, on computing resources provided by CRIANN (Centre Regional Informatique et d’Applications Numeriques de Normandie,
Normandy, France
Genetic Subdivision Curve and Surface Reconstruction
No full textIn this paper we employ a new genetic algorithm approach for CAD shape reconstruction, where a mathematical
shape representation is reconstructed from point data. We reconstruct planar subdivision curves and 3D subdivision
meshes from ordered input point data by fitting the corresponding subdivision control polygon or control mesh
respectively, from which the smooth subdivision limit surfaces can be derived. For the reconstruction of curves
the system estimates the number and position of control points required to approximate the curve closely. To
reconstruct subdivision surfaces from points, the system determines a sequence of CAD operations which is subject
to mutation in the course of a genetic optimization. We discuss implementation details of the proposed genetic
algorithms and demonstrate our approach on a number of example dat
Assessment culture of patient safety nursing students
No full textTato bakalářská práce se zabývá hodnocením kultury bezpečí pacientů studenty ošetřova-telství. Teoretická práce je zaměřená na přehled hodnotících a měřících nástrojů, které jsou využívání k hodnocení studenty ošetřovatelství. Je v práci představeno dvanáct do-tazníků. Dále jsou v práci představeny studie, které zahrnují využití těchto dotazníků. Ve výsledcích je demonstrováno využití jednotlivých dotazníků v klinické praxi.ObhájenoThis bachelor's thesis examines nursing students' assessment of patient safety culture. The theoretical work focuses on summarizing assessment and measurement tools that assess nursing students' patient safety culture. Twelve questionnaires are presented in the thesis. In addition, studies involving the use of these questionnaires are presented in the thesis. In the practical part, it is explained which questionnaires are used in practice and which ones are not
