Shai Avidan
Talk: Matching Data
I will present a number of algorithms for matching data in different settings.
The first involves matching 3D data point clouds in different settings (rigid, deformable, partial), I then extend this idea to matching two NeRFs (which are a popular 3D representation technique). The second setting involves matching points across images of different instances of the same class (say, different dogs, chairs, unicorns!). Lastly, I’ll show how to match text to image at a fine level of granularity.
BIO:
Shai Avidan received the PhD degree from the School of Computer Science, Hebrew University, Jerusalem, Israel, in 1999. He is currently a professor at the School of Electrical Engineering, Tel Aviv University. In between, he worked for Mobileye, Microsoft Research, Mitsubishi Electric Research Labs, and Adobe. He is interested in all things pixels.