Ariel Ephrat Research Scientist, Google DeepMind Email: arielephrat_at_google.com Google Scholar   |    Linkedin   |    Github

About Me

I am a Research Scientist at Google DeepMind, previously on the Perception Team at Google Research,
working on generative models for image and video.

I completed my PhD in 2018, working with Prof. Shmuel Peleg as part of the HUJI Machine Perception Group. My past research interests have included applications of computer vision using deep learning, particularly in the area of sight and sound-based video analysis.

Research & Product Highlights

Publications

	TokenVerse: Versatile Multi-concept Personalization in Token Modulation Space D. Garibi, S. Yadin, R. Paiss, O. Tov, S. Zada, A. Ephrat, T. Michaeli, I. Mosseri, T. Dekel arXiv, 2025 arXiv   |    webpage
	Still-Moving: Customized Video Generation without Customized Video Data H. Chefer, S. Zada, R. Paiss, A. Ephrat, O. Tov, M. Rubinstein, L. Wolf, T. Dekel, T. Michaeli, I. Mosseri SIGGRAPH Asia (Journal), 2024 arXiv   |    webpage   |    video
	Lumiere: A Space-Time Diffusion Model for Video Generation O. Bar-Tal, H. Chefer, O. Tov, C. Herrmann, R. Paiss, S. Zada, A. Ephrat, J. Hur, Y. Li, T. Michaeli, O. Wang, D. Sun, T. Dekel, I. Mosseri SIGGRAPH Asia, 2024 arXiv   |    webpage   |    video
	Teaching CLIP to Count to Ten R. Paiss, A. Ephrat, O. Tov, S. Zada, I. Mosseri, M. Irani, T. Dekel ICCV 2023 arXiv   |    webpage
	SpeedNet: Learning the Speediness in Videos S. Benaim, A. Ephrat, O. Lang, I. Mosseri, W.T. Freeman, M. Rubinstein, M. Irani, T. Dekel CVPR 2020, oral arXiv   |    webpage
	Neural Separation of Observed and Unobserved Distributions T. Halperin, A. Ephrat, Y. Hoshen ICML 2019 arXiv
	Dynamic Temporal Alignment of Speech to Lips T. Halperin*, A. Ephrat*, S. Peleg [* equal contribution] ICASSP 2019 arXiv   |    video   |    code
	Looking to Listen at the Cocktail Party: A Speaker-Independent Audio-Visual Model for Speech Separation A. Ephrat, I. Mosseri, O. Lang, T. Dekel, K. Wilson, A. Hassidim, W.T. Freeman, M. Rubinstein SIGGRAPH 2018 arXiv   |    webpage   |    video   |    Google Research Blog
	Seeing Through Noise: Visually Driven Speaker Separation and Enhancement A. Gabbay, A. Ephrat, T. Halperin and S. Peleg IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018 arXiv   |    webpage   |    video   |    code
	Patient-specific and global convolutional neural networks for robust automatic liver tumor delineation in follow-up CT studies R. Vivanti, L. Joskowicz, N. Lev Cohain, A. Ephrat and J. Sosna Medical & Biological Engineering & Computing, March 2018 article
	Improved Speech Reconstruction from Silent Video A. Ephrat*, T. Halperin* and S. Peleg [* equal contribution] ICCV 2017 Workshop on Computer Vision for Audio-Visual Media arXiv   |    webpage   |    video   |    code
	Vid2speech: Speech Reconstruction from Silent Video A. Ephrat and S. Peleg IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2017 pdf   |    webpage   |    code
	Compact CNN for Indexing Egocentric Videos Y. Poleg, A. Ephrat, C. Arora and S. Peleg IEEE Winter Conference on Applications of Computer Vision (WACV), 2016 pdf   |    webpage
	Automatic Liver Tumor Segmentation in Follow-Up CT Scans R. Vivanti, A. Ephrat, L. Joskowicz, O.A. Karaaslan, N. Lev Cohain and J. Sosna Proc. International Workshop on Patch-based Techniques in Medical Imaging, MICCAI, 2015 pdf

Teaching

HUJI — Teaching Assistant, Digital Image Processing, 2017-18 HUJI — Teaching Assistant, Undergraduate Engineering Projects and Workshop, 2015-18

Work

	Google DeepMind — Research Scientist, 2024 - Now
	Google Research — Software Engineer, 2018 - 2024 Google Research — Software Engineering Intern, 2017 Google — Software Engineering Intern, 2016
	Here a Story — Co-founder and Web Developer, 2013–2014

Research and product highlights inspired by Inbar Mosseri