Back from ICPR'12

In mid-November, I’ve been attending the ICPR 2012 conference in Tsukuba (Japan). It was a first time for me at a more “Pattern recognition”-oriented conference, with a slightly different community of attendees than ICIP or ICASSP, and also a different organization. I’ve been a bit suprised to have the feeling that the French community was bigger and more homogeneous than in these 2 conferences.

ICPR’12 facts

About the conference, I liked the fact that there were less sessions in parallel than ICIP or ICASSP1, so it’s actually a bit easier to attend various sessions. On the other hand, the poster sessions seemed to me less focused and it was not easy to understand the logic behind the board alignments. I won’t describe the overall program that you find here if you are interested.

If these numbers are of interest to you, there were 300+ papers accepted2 for oral presentation and a total of 900+ accepted papers. This makes for acceptance rates of 15% and 45% respectively, in line with previous editions.

I will keep a few highlights from the conference, like:

  • the plenary talk by Takeo Kanade about First Person Vision : wow, he knows how to keep an audience captive! A great speaker ! (I don’t need to talk about how research, because algorithms bearing his name are part of most Computer Vision text books and libraries…). It was also good to hear him talking about future projects, not just listing previous results and papers;
  • the invited talk by Nassir Navab about Patient and Process Specific Imaging and Visualization for computer assisted interventions: starting a talk with an excerpt from the Monty Pythons is always nice ;-) And I was pleasingly surprised to see how surgeons were starting to accept Augmented Reality technologies to help them in their work;
  • the full day tutorial by about __ Computer vision and image analysis in the study of master drawings and paintings__ by David Storke. It’s exciting to see Computer Vision and Computer Graphics technologies applied to arts analysis, David Storke is passionate about the topic, it was a great moment! You should really have a look at his publications, even just out of curiosity.

And I’m sure my advisor will endorse his advice:

If you’re a PhD student, take advantage of the fact of having no grant or whatever fundraising pressure by dedicating these few years to working on something original instead of tweaking some algorithm to beat the state-of-the-art by a few percents.

As with my previous conferences, I’m quite pleased by the tutorial I chose, and It has become one of my main motivations to actually attend this kind of event.

Our work at ICPR’12

Presenting our work on LBD reconstruction in the main conference hall (a really big one) was indeed a great experience. I apologize to the audience (if by any chance you read this blog) for being a bit confusing: as usual, I’ve tried to tell too many things in too short time :-(

I had some good feedback anyway from German and Austrian people, so maybe my style fits well with their habits or expectations (and these people were also more concerned about Computer Vision, feature extraction and what we presented).

Come on Barbie, let’s go party !

I was pleased of the stay in Tsukuba Science City. Lots of small houses, parks, a really nice place to live. And we could find some really good restaurants to eat okonomiyaki, katsudon or ramen!

Okonomiyaki, déc. 2012

Katsudon, déc. 2012

Not fast enough to take a picture of the ramen!

And since we were in Japan, the conference banquet ended Japanese style, i.e., a bit… abruptly. But, since we were indeed in Japan, we could pursue the social event in a karaoke bar -) 3.

Sake bottle at ICPR'12 banquet

And many thanks to the Austrian team for the time spent together the last two days, it was great meeting you!


  1. My fellows from the remote sensing field did actually complain that their conferences are even less parallelized. ^
  2. I believe the actual number is around 350, but I couldn’t find the exact number online. ^
  3. I keep my fingers crossed, hopefully no videos from that night will ever surface… ;-) ^

Related