19/06/2025
WHO: Sigal Raab, Tel-Aviv University
WHEN: Thursday, June 19th 2025 at 12:00
WHERE: BUILDING 503 (Computer Science) AUDITORIUM
Title: Synthesis and control of 3D motion
Abstract: Human motion is a fundamental attribute, underlying human actions, gestures, and behavior, with applications spanning autonomous driving, security, and entertainment. In my talk, I will present my research on novel approaches to 3D motion synthesis and control using deep neural networks. I'll discuss fundamental animation challenges including temporal consistency and skeletal structure learning. The presentation will explore how diffusion models can be adapted for 3D motion, enabling the synthesis of natural and diverse sequences. Special attention will be given to techniques for learning from single motion examples and extending these approaches to animal motion.
Short Bio: Sigal Raab is a computer graphics and vision researcher pursuing her PhD at Tel Aviv University under the guidance of Prof. Daniel Cohen-Or, a path she embarked on after years developing cutting-edge products and algorithms in the high-tech industry. A graduate of Bar Ilan University, where she earned both her bachelor's and master's degrees summa cum laude, Raab specializes in 3D motion synthesis and control through innovative deep learning approaches. Her research tackles the complex challenges of motion analysis, developing solutions with wide-ranging applications across fields including medical simulations, entertainment media, and beyond.
