Computer Vision News - October 2018

Angjoo Kanazawa Daily Wednesday Angjoo Kanazawa is a Berkeley Artificial Intelligence Research Lab (BAIR) postdoc at the University of California Berkeley advised by Jitendra Malik, Alyosha Efros and Trevor Darrell. She completed her Ph.D. in Computer Science at the University of Maryland (UMD), College Park with her advisor David Jacobs. Prior to UMD, she spent four years at New York University (NYU) where completed her B.A. in Mathematics and Computer Science, with Rob Fergus as a teacher. Angjoo, can you tell us about your work? I study single-view 3D reconstruction of non-rigid objects like humans and animals, and how we can understand their state, what they are doing from a single or monocular camera, and hopefully use this to imitate or understand human behavior more. How did you get to this topic? Originally, I was doing classification. It was also okay, but I was kind of tired of getting numbers, and I really like cats. I thought, “Okay, I might as well make good use of my time .” I want to make a reconstruction of cats. A very legitimate ambition. [ both laugh ] Humans are important, but yeah… When you get your Nobel prize, you will say it all started with your love for cats. Do you have cats? I have two. I have an Instagram account for them . For each of them? No, just both of them. Aren’t they jealous of one another? [ both laugh ] You were not born in Berkeley, right? I was born in Japan. Where in Japan? Osaka. Kobe. Women in Computer Vision 26 Angjoo (left) with Michael Black , David Jacobs and JitendraMalik