Aleksander Holynski

Aleksander Holynski is an Assistant Professor of Computer Science at Columbia University. He received his Ph.D. from the University of Washington and his B.S. from the University of Illinois at Urbana-Champaign, following postdoctoral research at UC Berkeley (BAIR) and a research scientist role at Google DeepMind. His research focuses on computer vision and graphics, specializing in generative 3D scene reconstruction, dynamic view synthesis, and image-to-video synthesis. He is the recipient of the ICCV 2023 Best Student Paper Award for "Tracking Everything Everywhere All at Once" and the CVPR 2024 Best Paper Award for "Generative Image Dynamics." His recent work includes continuous 3D perception models with persistent state and test-time training for linear-time reconstruction, with collaborations spanning Google DeepMind and UC Berkeley.