Irfan Essa

Irfan Essa is a Professor at the School of Interactive Computing, Georgia Institute of Technology, and serves as Senior Associate Dean in the College of Computing. He leads the Interdisciplinary Research Center for Machine Learning at Georgia Tech (ML@GT) and holds a Senior Staff Research Scientist position at Google Inc. His academic journey includes a Ph.D. from MIT (1994) and prior research faculty roles at MIT Media Lab (1988-1996).

• Research Focus: Computer Vision, Machine Learning, Robotics, Computational Journalism, and Human-Computer Interaction.
• Leadership: Co-founded ML@GT, Adjunct Faculty at Carnegie Mellon's Robotics Institute.

Essa’s research explores Generative AI, Computational Video, Wearable Computing, and Vision-Language Models. His work impacts areas like Autonomous Systems, Video Analysis, and Social Computing. Recent projects include end-to-end trainable frameworks for Image Retargeting (HALO) and Video Generation (VideoPoet), emphasizing content-structure preservation and zero-shot capabilities.

His 15 most recent publications (2023-2025) span CVPR, NeurIPS, ICML, and ECCV, focusing on Generative Media, Diffusion Models, and Human Activity Recognition. Notable contributions include Text-Free Diffusion, Wearable-Based Activity Recognition, and Layered Transformation Architectures.

• Awards: NSF CAREER, IEEE Fellow, Best Paper Awards at ICML 2024 and ICLR 2023.
• Students: Advisees include Varun Agrawal, Aneeq Zia, Unaiza Ahsan, and Daniel Castro Chin.
• Labs/Teams: Affiliated with ML@GT, Georgia Tech's Center for Experimental Research in Computer Systems (CERCS), and Google Research.