Sergey Tulyakov

Sergey Tulyakov is the Director of Research at Snap Inc., leading the Creative Vision team. His work focuses on enhancing creator capabilities through computer vision, machine learning, and generative AI, with applications in 2D/3D/4D video generation, editing, and personalization. He pioneered video generation frameworks like MoCoGAN and First Order Motion Model, and has been recognized for BEST IN SHOW AWARD at SIGGRAPH Real-Time Live! 2020.

• PhD (2012-2017): University of Trento, Italy
• MSc (2010): Belorusian State University of Informatics and Radioelectronics
• B.Eng (2009): Belorusian State University of Informatics and Radioelectronics

His research interests span computer vision, generative models, 3D reconstruction, and personalization, with a focus on making large models efficient and mobile-compatible. Recent publications highlight advancements in 4D video generation, text-guided 3D composition, and lightweight architectures.

Key scientific awards include the SIGGRAPH Real-Time Live! 2020 Best in Show for Interactive Video Stylization. He has also served on technical program committees for top-tier conferences like CVPR, ICCV, SIGGRAPH, and NeurIPS since 2022.

His team organizes tutorials and keynotes, including courses on Deep Generative Models and Efficient Neural Networks. While no direct student names are listed, his collaborative work spans 60+ top-tier publications.