Phillip Isola
MIT associate professor in EECS; first author of pix2pix and co-author of Foundations of Computer Vision
- Associate Professor in EECS; member of CSAIL, Massachusetts Institute of Technology (EECS and CSAIL)12
Phillip Isola is an associate professor in EECS at MIT and a member of CSAIL, where Isola's group studies fundamental principles of intelligence, including how different models come to learn similar representations. Isola is first author of "Image-to-Image Translation with Conditional Adversarial Networks" (the pix2pix paper), whose code is released under a BSD-style license. Isola co-wrote the textbook Foundations of Computer Vision with Antonio Torralba and William Freeman, which is freely readable online, and is listed as an instructor of the Fall 2024 offering of MIT's grad-level deep learning course 6.7960, published on MIT OpenCourseWare. Isola's MIT bio notes earlier time at UC Berkeley, OpenAI, and Google Research.123456
Work
pix2pix (Image-to-Image Translation with Conditional Adversarial Networks) (external site: github.com)Open-source code
First author, with Jun-Yan Zhu, Tinghui Zhou, and Alexei A. Efros, of the 2016 paper on conditional adversarial networks for image-to-image translation; the original code repository is released under a BSD-style license naming Isola and Jun-Yan Zhu as copyright holders.34
Foundations of Computer Vision (external site: visionbook.mit.edu)Published research
Co-author, with Antonio Torralba and William Freeman, of a computer vision textbook published by The MIT Press and freely available in an online form.51
MIT 6.7960 Deep Learning (Fall 2024) (external site: ocw.mit.edu)Published research
Listed as an instructor, with Sara Beery and Jeremy Bernstein, of the graduate deep learning course published on MIT OpenCourseWare.61
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.