SIGGRAPH Time Test Award Revealed: This Research Bet on Physical AI 10 Years Ahead

SIGGRAPH 2026 named a 2016 motion synthesis paper from Taku Komura's University of Hong Kong team as a Test-of-Time Award winner.

The work learned human motion representations from large-scale movement data to generate and edit natural character animation.

It became a key foundation for later research in motion generation, scene interaction, and multimodal embodied intelligence.

The paper is seen as an early driver of physical AI and embodied AI efforts that model interactions between bodies, objects, and environments.