HiPHI: A Large-Scale Benchmark for High-Precision Human Motion and Object Interaction

<img src="https://spectrum.ieee.org/media-library/noitom-robotics-logo-with-stylized-nr-monogram-on-transparent-background.png?id=68382308&width=980"/><br/><br/><p><span>This White Paper gives robotics researchers and engineers an overview of a new large-scale motion capture dataset built to close the data gap limiting humanoid robot learning. It also shows how policies trained on the dataset transfer to a real humanoid robot.</span></p><p><strong>What you will learn about: </strong></p><ul><li>Why humanoid robot learning, a central problem in embodied AI and Physical AI, needs data that internet video and existing motion capture datasets cannot provide. </li><li><span><span>How </span><span>FrameNet</span><span>, a linguistic framework for human action, can guide motion capture collection to systematically cover a broad range of whole-body motion.</span></span> </li><li><span><span>Why synchronized object trajectories and meshes make human-object interaction data useful for teaching </span><span>robots</span> real-world tasks such as carrying, pushing, and pulling.</span> </li><li><span>How reinforcement learning policies trained on this motion capture data improve with scale, and how sim-to-real transfer carries them onto a physical humanoid robot.</span></li></ul><div><a href="https://content.knowledgehub.wiley.com/hiphi-a-large-scale-benchmark-for-high-precision-human-motion-and-object-interaction/" target="_blank">Download this free whitepaper now!</a></div>