EuroVis 2011

I’ve written before about EuroVis, the most important European scientific conference on visualisation. In 2009, it took place in Berlin, in 2010 it was in Bordeaux, and, an a surprise non-twist of alliteration, the 2011 edition was held in Bergen, Norway. With 216 attendees and a practically perfect organization, this year’s edition has been described as the biggest and the best EuroVis ever. In a bid to save some time (I still owe you a mega-edition of the Weekly (actually Monthly) Head Voices), I’m going to give my biased account in bullet-list form:

• On Tuesday evening, we were welcomed by the very charismatic vice mayor of Bergen in the Tårnsalen of the Lysverket building of the Bergen Art Museum. It seems the photo I took of the inside of the art deco tower, built in 1938, is quite a popular shot. The food was divine, thank you very much.

  

• The next morning, during the conference opening, the following Bergen (rainiest city in Europe) joke was told: A visitor asks a local boy in exasperation “Does it rain like this all the time?” and the little boy answers “I don’t know, I’m only 12 years old!”.

• The conference keynote was given by Scott McCloud, American cartoonist and comic theorist, on aspects of visual communication. This was most probably the best presentation I’ve ever had the privilege of experiencing. Besides brilliant oratorship, his slides are somehow more a visual stream of consciousness affair than discrete quanta of information. When I grow up, I’m going to present like that.

As per usual, I get to award the Weekly Head Voices Best Paper awards, and they go to the following papers:

• A Shader Framework for Rapid Prototyping of GPU-Based Volume Rendering by Christian Rieder, Stephan Palmer, Florian Link and Horst K. Hahn. Rieder and his colleagues have constructed a full GPU-based volume rendering pipeline in MeVisLab of which the various shader based components are modifiable at runtime. This means that you can prototype your GPU-based volume rendering ideas in no time flat!
• Curve Density Estimates by Ove Daae Lampe and Helwig Hauser. Back to basics and really important work on the effective visualisation of complex curves at any resolution, with smooth scaling between levels.
• A Gradient-Based Comparison Measure for Visual Analysis of Multifield Data by Suthambhara Nagaraj, Vijay Natarajan and Ravi S. Nanjundiah. Another back-to-basics paper in which the authors show how to find the agreement between hundreds of scalar fields and visualise this agreement, thus enabling comparison.

The slightly less prestigious EuroVis 2011 Best Paper awards went to:

1. Uncertainty-Aware Exploration of Continuous Parameter Spaces Using Multivariate Prediction  by Wolfgang Berger, Harald Piringer, Peter Filzmoser, Eduard Gröller. I was unfortunately in the other session, but was told by numerous colleagues that this was indeed an award-winning presentation as well.
2. A User Study of Visualization Effectiveness Using EEG and Cognitive Load by Erik Anderson, Kristin Potter, Laura Matzen, Jason Shepherd, Gilbert Preston, Claudio Silva. This was presented in the Evaluation session which I had the privilege of chairing. It is indeed a very compelling idea to measure the effectiveness of a visualisation through cognitive load and this paper documents the first very important steps in this direction.
3. A Gradient-Based Comparison Measure for Visual Analysis of Multifield Data by Suthambhara Nagaraj, Vijay Natarajan and Ravi S. Nanjundiah. This was also amongst the more prestigious WHV best paper award winners, see above!

The rest of the conference featured the following bullets:

• During the social event on Thursday evening, Frits Post (my boss), was elevated to the rank of Eurographics Honorary Fellow, recognizing his service to and standing in the visualisation community. Including this newest addition, there are only five (5!) EG Honorary Fellows in the world today. I am very proud!
• During the first session of the morning after the social event, I had the exquisite privilege of presenting the work of my Brazilian colleagues: Piecewise Laplacian-based Projection for Interactive Data Exploration and Organization by Fernando V. Paulovich, Danilo M. Eler, Jorge Poco, Charl P. Botha, Rosane Minghim, Luis G. Nonato. I really do like presenting at events like these, and it’s been a while. Do read and cite the paper, it documents a practical way of reducing any set of high-dimensional data points to the visual space, and enabling interaction with those points on the visual space!
• The capstone of the conference was presented by the legendary Prof. E. Gröller, also known by the whole community as Meister. In typical style, the title of his talk was only announced during the talk itself. The title was The Haunted Swamps of Heuristics. In this philosophical and visionary contribution, it was argued that algorithms and parameters are too deeply intertwined to focus only on the former, but that it was more important to study, in detail, the exact behaviour of the latter. More broadly speaking, we need to accept the fact that there is a great deal of uncertainty also in the parameter spaces of our algorithms, but that this uncertainty can and should be dealt with correctly.

That thought-provoking capstone and this blog post will share the same concluding quote:

Doubt is not a pleasant condition, but certainty is absurd. — Voltaire / Gröller

I hope to see you in the comments below! You could also opt to click on my shiny new +1 button, or my slightly older but no less shiny retweet or facebook share buttons.