The Top Ten Lab Directors currently are:
Points Name & Lab ========================================================= 252.57 Antoine McNamara, The Inverse Kinematics 223.00 Brian Curless, Ankle twisters 221.00 Brett Allen, The Search for the Holy Grail 206.00 C. Karen Liu, Nutcrackers 199.00 Xavier Granier, HWG 192.00 Ville Miettinen and Timo Aila, Puoli koppaa 188.57 Aaron Hertzmann, Battle Purple Sea Urchin! 177.57 Hugues Hoppe, Power in numbers 177.00 Piotr Piastucki, Nightmare Labs 177.00 Eitan Grinspun, The Shape Operators
out of 96 contestants. The previous year's winner, Aaron Hertzmann, is in seventh place at this point. However, sketches remain to be judged, and each sketch author is worth 20 points, or if touching two or more edges, 20 vertices.
So what good's a database if you can't mine it for random stuff? Here's a breakdown of the most popular words used in SIGGRAPH paper titles this year:
Title words, by popularity ========================== 20 for 18 of 15 and 10 A 8 Using 7 on 7 with 7 the 6 Hardware 6 Graphics 5 Texture 5 Synthesis 5 From 5 Motion 4 Light 4 Image 4 Human 4 in 4 Video 4 Rendering 4 Interactive 4 Shadow 3 Skin 3 Modeling 3 Maps 3 Complex 3 Mesh 3 Character 3 Parameterization 3 Meshes 3 3DSo one sweet-spot title for a paper for next year is: "Motion Synthesis from Texture using Graphics Hardware". Hmmm, that actually sounds tantalizing... Other, second-tier titles include (using the 4's), "Rendering Interactive Human Shadows in Video Light," and (using the 3's) "Complex Skin Modeling Maps for 3D Character Mesh Parameterization." These papers practically write themselves once you have their titles.
Speaking of popularity, here's which researchers were chosen by the most lab directors:
Authors, by popularity on teams =============================== 13 Zoran Popovic 13 Michael D. McCool 12 Jovan Popovic 12 Wolfgang Heidrich 11 Frédo Durand 11 Peter-Pike Sloan 9 Ken Perlin 9 Paul E. Debevec 8 Henrik Wann Jensen 7 Jan Kautz 7 Pat HanrahanWere they worth it? Here are the ten highest payoffs for researchers that you could hire:
### Value Cost Efficiency Name =============================== 602 85.00 100 0.8500 Zoran Popovic 281 75.00 200 0.3750 Jos Stam 63 72.00 110 0.6545 Bruno Lévy 365 67.00 80 0.8375 Mathieu Desbrun 181 62.00 90 0.6889 Hans-Peter Seidel 428 60.57 260 0.2330 Pat Hanrahan 408 60.00 10 6.0000 Nancy S. Pollard 90 60.00 40 1.5000 Daniel Cohen-Or 444 55.00 140 0.3929 Peter Schröder 449 45.00 20 2.2500 Philip Dutré
So on a pure points basis, only two of the popular choices paid off this year, Zoran and Pat, both of whom are pricey. Some of the other popular choices scored points, but did not yield the greatest payoffs.
Now let's look at efficiency, i.e. most bang for the buck. Here is everyone with an efficiency of 2.0 or more:
### Value Cost Efficiency Name =============================== 408 60.00 10 6.0000 Nancy S. Pollard 437 30.00 10 3.0000 Paul S. A. Reitsma 450 15.00 5 3.0000 Philip L. Davidson 470 15.00 5 3.0000 Robert D. Kalnins 449 45.00 20 2.2500 Philip Dutré 286 20.00 10 2.0000 Jovan Popovic 546 20.00 10 2.0000 Tomoyuki Nishita 588 20.00 10 2.0000 Ren Ng 594 20.00 10 2.0000 Yoshinori DobashiHats off to Nancy Pollard, who was well ahead of the pack here and also appears on the highest payoff list. She was chosen by two teams. As usual, few of these were the most popular choices (actually, just Jovan), since the most cost-effective researchers have a low cost and so usually are not as well-known (yet).
And that's all until the sketches are announced. If you want to check your results or have some strange craving to actually see what papers were accepted, check Tim Rowley's great web page. I've also put up the alphabetical listing of past researchers and how many points each is worth this year.
In the meantime, for a walk down memory lane, recalling the halycon days of your youth, don't bother with these links: