Konversatorium on Friday, June 26, 2020 - 10:30

Slots for Talks still available! Please contact the KV administration.
Friday, June 26, 2020 - 10:30
Zoom Meeting

Accelerating Ray Tracing Using FPGAs (DAEV)

Alexander Reznicek (Inst. 193-02 CG)
20 + 20
Hiroyuki Sakai

The synthesis of an image from a scene stored on a computer is called rendering, which is able to deliver photo-realistic results, e.g., by using specific variants of the class of ray tracing
algorithms. However, these variants (e.g., path tracing) possess a stochastic characteristic which results in a high computational expense. This is explained by the nature of stochastic algorithms, which use a high number of samples to compute a result—in case of ray tracing, these samples manifest in a high number of rays needed for a complete rendering.

One possibility to accelerate ray tracing—no matter if using a stochastic or simpler variants—is the use of customized hardware. FPGRay is such an approach, which combines the use of customized hardware with the software of an off-the-shelf PC to a hybrid solution. This allows increasing the efficiency by specialized hardware and delivers a sustainability in case of changing algorithms at the same time.

The results point towards a possible efficiency gain. Unfortunately, in the scope of this thesis this was not realizable and the specific implementation showed a lower efficiency compared to the software implementation. Nevertheless, the possibility to achieve a higher efficiency with this approach by indicating FPGRay’s potential could be shown.

Multilevel Layout of Metabolite Networks (DAAV)

Stefanie Prast (Inst. 193-02)
10 + 10
Hsiang-Yun WU

In life sciences, a series of successive reactions within a cell are linked together to form metabolic pathways. These can be visualized as graphs and networks, but they are complex
and constantly changing. An automated visualization and layout
approach is therefore necessary. Graphical notations to describe these networks are established in life sciences, like the SBGN (System Biology Graph Notation).
Available software tools support the visual encoding of SBGN, but provide only standard means for automatic arrangement, e.g. hierarchical, circular, or tree layouts, these do not satisfy the
layout rules of SBGN. Constraint based layouts offer means to model such layout rules, but
current approaches only focus on subsets or simplifications of SBGN, e.g. compound structures are almost never taken into account during the layout process.
Providing suitable means to visualize SBGN while considering its semantics during graph layout is therefore still an open topic. In this thesis we plan to develop a multilevel layout strategy that
takes the graphical notation of life sciences into account and visualizes a metabolic pathway's data flow and compound structure, while minimizing the complexity of the network by motif simplification and semantic zooming.