Sebastian Starke receives the EUROGRAPHICS Young Researcher Award 2025. Sebastian obtained his PhD from the University of Edinburgh under the supervision of Taku Komura. He is now a research scientist at Meta Reality Labs.
Sebastian has made significant contributions to motion synthesis and character animation methods using deep learning techniques. His research in character animation fuses motion control with deep learning to create responsive and lifelike digital characters. In his research, Sebastian extends the phase concept into complex human-scene interactions, such as basketball playing, boxing and our four-legged friends. His DeepPhase framework introduces an end-to-end neural architecture that learns a compact, representative phase space directly from raw motion capture data. This approach not only unifies the existing phase-based representations, but also elegantly handles the nuances of diverse motion patterns, ensuring natural and fluid animation synthesis. Separately, his codebook matching algorithm addresses the inherent ambiguities of control signals -such as those from VR devices- by aligning and matching latent categorical probability distributions. By explicitly sampling from the distribution, the technique results in high-fidelity and responsive control systems that are pivotal for immersive applications in embodiment in metaverse and beyond.
The work of Sebastian Starke is published in the top tier conferences and journals of computer graphics and has been widely cited. His work received several honors, such as best paper awards at SIGGRAPH and Pacific Graphics, as well as the Symposium on Computer Animation (SCA) Best PhD Dissertation award (2023). Sebastian’s innovative contributions to the field of character animation significantly advance interactive applications such as gaming, virtual reality and robotics.
EUROGRAPHICS is extremely pleased to recognize Sebastian Starke with the 2025 Young Researcher Award in recognition of his outstanding contributions to Computer Graphics in the area of character animation and motion synthesis.