Erwan Leria ‑ early stage researcher within VR

Wed 12 Oct 2022 16:17

Erwan Leria works as an early stage researcher at Tampere University, Finland, and Mid Sweden University. His research topic is “Real-time View Interpolation with Distributed Heterogeneous Computing“.

Erwan Leria
Erwan Leria early stage researcher at Tampere University, Finland, and Mid Sweden University.

Tell me about yourself, your background?

I grew up in Bordeaux, in the southwest of France. My high school and my first university were also in Bordeaux. I completed first a Bachelor of Computer Science in 2018. Then, I moved to Marseille where I did my Master’s degree in Geometry and Computer Graphics.

During my studies, I worked twice at the French National Institute for Research in Computer Science and Automation (INRIA) as a Research Intern. The first time was in Bordeaux, for my Bachelor’s thesis, where the topic was about performance simulation of replayed tasks graphs for scheduling strategies. The other one was at Grenoble, for my Master’s thesis, where I worked on real-time volumetric rendering of nebulae generated by multifractal Perlin noise.

What have you done so far as an PhD-student?

I work within the Virtual reality and Graphics Architecture group of Tampere University, my research topic focuses on distributed and parallel real-time multiview path tracing for light field displays.

I have been looking through the previous work done by peers in my research topic. A few works have been done in this field. Therefore, I created a model to formally describe the interactions between the synthetic generation of multiple image views from 3D scenes and a parallel platform composed of multiple compute devices such as GPUs. Now the goal is to extend this model in order to increase the scalability of rendering applications based on light field technologies. Quantitative measurements will be useful to measure the efficiency of my approach and to consider also its limitations.

Why did you apply to the Plenoptima project?

I was looking for a research position which deals with parallel heterogeneous computing and real-time rendering. Parallel real-time rendering on heterogeneous architectures is a subject in which I have been interested since I was a kid. The combination of these two topics is not really studied in France, as far as I know, so I was ready to look for opportunities outside of France. My only requirement was to stay in the European Union area, which would ease some administrative requirements. It is also better to be closer to my family and avoid very long-distance trips for holidays. The early-stage researcher positions of the Plenoptima project were available a few months after I graduated in 2020. The topic of the ESR4 was exactly what I was looking for, and the main location for this position was in Finland, which was in the area I was considering. Moreover, being part of the Plenoptima project is the opportunity to meet new people and to be part of a project which has a concrete direction.

What are your thoughts about the upcoming visit as a PhD-student at Mid Sweden University?

In the scope of my doctoral project, I must spend six months in Mid-Sweden University. I plan to come this January. This kind of visit tends to foster mobility within European countries and is an opportunity for me to enlarge my professional network. Sundsvall seems to be a quiet and cute place. I am pretty sure that there are plenty of things to do and to visit there and more globally in the country in general which I am not familiar with.

It has been a year now that we are planning this short secondment with my supervisors from both Tampere and Mid Sweden University. I am convinced that during my visit I will progress in my research work where I will focus on distributed computing for multiview path tracing. I know that the quality of life is good in Sweden and even better in Sundsvall, this is the perfect match for me to only focus on my research without being worried by other stuff.

What kind of expectations do you have for your research project and the upcoming network training?

I would be glad to contribute new ideas and perspectives to the scientific community. I want to bring new topics and problems to the table in order to prepare for next generations of VR/AR experience in real-time. I expect my research to give an understanding of the challenges to build new eco-systems for different environments where real-time parallel rendering applications can evolve. This would possibly lead to reconsidering the hardware and low-level implementations of the current parallel architectures to make them fit real world applications that are computationally demanding. Overall, I am sure that this work will contribute to society, enabling new ways to visualize and interact with virtual worlds and data that surround us.

As I am permanently seeking new knowledge, I expect the Plenoptima training network to be the opportunity to develop new skills and to learn more about new things while I’m working on a topic that I’m interested in.

Contact

Erwan Leria

Doktorand|Doctoral Student


Recommended

The page was updated 10/12/2022