Dr. Jona Ballé

Biography

Jona Ballé is an Associate Professor at NYU Tandon School of Engineering, specializing in visual media compression and machine learning. His research focuses on developing efficient representations for various types of visual media, including images and video, as well as novel modalities like augmented and virtual reality. Ballé's work heavily relies on machine learning and end-to-end optimization, aimed at improving compression algorithms by adapting to data. He is dedicated to understanding how humans perceive visual scenes to enhance compression results. He defended his master’s and doctoral theses in signal processing and image compression under the supervision of Jens-Rainer Ohm at RWTH Aachen University from 2007 to 2012. He has collaborated with prominent figures, including Javier Portilla at CSIC in Madrid and worked at New York University’s Center for Neural Science under Eero P. Simoncelli, where he studied the relationship between perception and image statistics. Ballé pioneered the use of machine learning for end-to-end optimized image compression, which significantly contributed to the JPEG AI standard, finalized for 2025. Between 2017 and 2024, he deepened his ties with the industry as a Research Scientist at Google, while continuing his line of research. He serves as a reviewer for top-tier publications in machine learning and image processing, including NeurIPS, ICLR, ICML, and IEEE Transactions. He is also an active co-organizer of the annual Challenge on Learned Image Compression (CLIC) since 2018 and a program committee member for the Data Compression Conference (DCC) in 2022.

Research Interests

Experience