Computer Graphics & Visual Computing (CGVC) 2022

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/2633209

Browse

Browsing Computer Graphics & Visual Computing (CGVC) 2022 by Author "Maddock, Steve"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Augmented Reality for Safety Zones in Human-Robot Collaboration
    (The Eurographics Association, 2022) Cogurcu, Yunus Emre; Douthwaite, James A.; Maddock, Steve; Peter Vangorp; Martin J. Turner
    Worker productivity in manufacturing could be increased by reducing the distance between robots and humans in human-robot collaboration (HRC). However, physical cages generally limit this interaction. We use Augmented Reality (AR) to visualise virtual safety zones on a real robot arm, thereby replacing the physical cages and bringing humans and robots closer together. We demonstrate this with a collaborative pick and place application that makes use of a Universal Robots 10 (UR10) robot arm and a Microsoft HoloLens 2 for control and visualisation. This mimics a real task in an industrial robot cell. The virtual safety zone sizes are based on ISO standards for HRC. However, we are the first to also consider hardware and network latencies in the calculations of the virtual safety zone sizes.
  • No Thumbnail Available
    Item
    Depth-aware Neural Style Transfer using Instance Normalization
    (The Eurographics Association, 2022) Ioannou, Eleftherios; Maddock, Steve; Peter Vangorp; Martin J. Turner
    Neural Style Transfer (NST) is concerned with the artistic stylization of visual media. It can be described as the process of transferring the style of an artistic image onto an ordinary photograph. Recently, a number of studies have considered the enhancement of the depth-preserving capabilities of the NST algorithms to address the undesired effects that occur when the input content images include numerous objects at various depths. Our approach uses a deep residual convolutional network with instance normalization layers that utilizes an advanced depth prediction network to integrate depth preservation as an additional loss function to content and style. We demonstrate results that are effective in retaining the depth and global structure of content images. Three different evaluation processes show that our system is capable of preserving the structure of the stylized results while exhibiting style-capture capabilities and aesthetic qualities comparable or superior to state-of-the-art methods. Project page: https://ioannoue.github.io/depth-aware-nst-using-in.html.