The detection of pedestrians plays an essential part in the development of automated driver assistance systems. Many of the currently available datasets for pedestrian detection focus on urban environments. State-of-the-art neural networks trained on these datasets struggle in generalizing their ...

This paper presents a new approach for the vectorization of stylized images using intermediate data representations to interface image stylization and vectorization techniques. It enables the combination of efficient GPU-based implementations of interactive image stylization techniques and the advantages&#...

Airborne light detection and ranging (LIDAR)- based bathymetry is a highly specialized field within the widely known and used geoscientific surveying technology based on green spectrum lasers. Green light can penetrate shallow water bodies such that river and lake beds can be surveyed.&...

Since RGB-D sensors became massively popular and are used in a wide range of applications, depth data compression became an important research topic. Live-streaming of depth data requires quick compression and decompression. Accurate preservation of information is crucial in order to prevent...

The rectilinear Perspective projection produces natural-looking results on the condition that the degree of field of view (FoV) is narrow, as raising it causes an exponential increase in visual distortion. Curvilinear perspective projection methods that counter this issue exist in photography,&#x...

In computer graphics, the generalized barycentric coordinates (GBC) are often used for image deformation. To manipulate an input image using a cage based image deformation method, we usually have to consider a source polygon with a triangulation; but defining the triangulation of the&#x...

Nowadays, when one needs a system for image recognition, it is mostly a matter of finding pre-trained CNN and, sometimes, adding additional training based on transferred knowledge. Accurate 6-DOF object localization in the image is a more laborious task and requires more complex tr...

Automatic extraction of building roofs from remote sensing data is important for many applications including 3D city modeling, urban planning, disaster management, and simulations. In this paper, we propose an automatic workflow for roof reconstruction by polygonal models from classified high-den...

We demonstrate the acceleration potential of the Chebyshev semi-iterative approach for fluid simulations in Projective Dynamics. The Chebyshev approach has been successfully tested for deformable bodies, where the dynamical system behaves relatively linearly, even though Projective Dynamics, in general,&#x...

This paper reports on a service-based approach to enable real-time stream-processing of high-resolution videos. It presents a concept for integrating black-box image and video processing operations into a streaming framework. It further describes approaches to optimize data flow between the proce...

We present an experimental setup to capture photon distributions from a liquid scattering medium and to reconstruct the flux density throughout the medium. The capture mechanism moves an occluder through the medium, which is illuminated by a laser. For each position of the occluder...

Indoor reconstruction using depth camera algorithms (e.g., InfiniTAMv3) is becoming increasingly popular. Simple reconstruction methods solely use the frames of the depth camera, leaving any imagery from the adjunct RGB camera untouched. Recent approaches also incorporate color camera information to&#...

In this paper, we propose a new semi-fragile watermarking scheme in the frequency domain for surveillance videos authentication. Our system starts operating by generating a binary watermark based on a novel watermark construction process. This latter combines Speeded Up Robust Features (SURF...

Deep learning approaches suffer from the so called interpretability problem and can therefore be very hard to visualise. Embedded Prototype Subspace Classifiers is one attempt in the field of explainable AI, which is both fast and efficient since it does not require repeated learni...

With 360 imaging devices becoming widely accessible, omnidirectional content has gained popularity in multiple fields. The ability to estimate depth from a single omnidirectional image can benefit applications such as robotics navigation and virtual reality. However, existing depth estimation approach...

The assessment of the accuracy and stability of model-based tracking approaches in the literature is a challenging task. Many of the presented methods lack a detailed evaluation or comparison to accurate ground truth data. Considering real world applications, it is hard to estimate ...

In the paper an algorithm for description of greyscale objects extracted from images is applied for recognition of human red blood cells visible on digital microscopic images. This is a part of an approach for automatic (or semiautomatic) diagnosis of selected diseases based on...

Face hallucination is a type of super resolution that restores very low resolution (8 8 pixel) to high resolution (128 128 pixel) face images. Since unique facial features caused by age, e.g.wrinkles, are ignored during restoration, restored face images can be somewhat dissimilar t...

3D mesh unfolding transforms a 3D mesh model into one or multiple 2D planar patches. The technique is widely used to fabricate papercrafts, where 3D objects can be reconstructed from printed paper or paper-like materials. The applicability, visual quality, and stability of such pap...

With the steady progress of Deep Learning (DL), powerful tools are now present for sophisticated segmentation tasks. Nevertheless, the generally very high demand for training data and precise reference segmentations often cannot be met in medical domains when processing small and individual&...