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

DevFeed: [HiPHI: A Large-Scale Benchmark for High-Precision Human Motion and Object Interaction](<https://devfeed.tech/articles/hiphi-a-large-scale-benchmark-for-high-precision-human-motion-and-object-interaction-66717.md>)

Original publisher: [Read original article](<https://content.knowledgehub.wiley.com/hiphi-a-large-scale-benchmark-for-high-precision-human-motion-and-object-interaction/>)

Author: Noitom Robotics

Published: 2026-10-07T13:49:48Z

Content type: article

Language: en

Sources: [IEEE Spectrum](<https://devfeed.tech/sources/ieee-spectrum.md>)

Topics: [Humanoid Robots](<https://devfeed.tech/topics/humanoid-robots.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Robot Manipulation](<https://devfeed.tech/topics/robot-manipulation.md>), [ImageNet](<https://devfeed.tech/topics/imagenet.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmark-dataset](<https://devfeed.tech/tags/benchmark-dataset.md>), [embodied-ai](<https://devfeed.tech/tags/embodied-ai.md>), [framenet](<https://devfeed.tech/tags/framenet.md>), [human-motion-dataset](<https://devfeed.tech/tags/human-motion-dataset.md>), [human-object-interaction-dataset](<https://devfeed.tech/tags/human-object-interaction-dataset.md>), [humanobject-interaction](<https://devfeed.tech/tags/humanobject-interaction.md>), [humanoid](<https://devfeed.tech/tags/humanoid.md>), [humanoid-robot-learning](<https://devfeed.tech/tags/humanoid-robot-learning.md>), [humanoid-robots](<https://devfeed.tech/tags/humanoid-robots.md>), [motion-capture](<https://devfeed.tech/tags/motion-capture.md>), [motion-space-coverage](<https://devfeed.tech/tags/motion-space-coverage.md>), [motion-tracking](<https://devfeed.tech/tags/motion-tracking.md>), [object-trajectory-data-supplementary-keywords-client-provided-motion-capture](<https://devfeed.tech/tags/object-trajectory-data-supplementary-keywords-client-provided-motion-capture.md>), [optical-motion-capture](<https://devfeed.tech/tags/optical-motion-capture.md>), [paper](<https://devfeed.tech/tags/paper.md>), [physical-ai](<https://devfeed.tech/tags/physical-ai.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [robot-policy-learning](<https://devfeed.tech/tags/robot-policy-learning.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [sim-to-real-transfer](<https://devfeed.tech/tags/sim-to-real-transfer.md>), [sub-millimeter-tracking](<https://devfeed.tech/tags/sub-millimeter-tracking.md>), [type-whitepaper](<https://devfeed.tech/tags/type-whitepaper.md>), [whole-body-control](<https://devfeed.tech/tags/whole-body-control.md>), [whole-body-motion](<https://devfeed.tech/tags/whole-body-motion.md>)

## AI overview

This white paper describes HiPHI, a large-scale motion capture dataset for humanoid robot learning. It covers whole-body motion and human-object interactions, uses FrameNet to organize action coverage, and reports benchmark results and transfer of trained policies to a physical Unitree G1 robot.

## Source excerpt

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. What you will learn about: 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. How FrameNet, a linguistic framework for human action, can guide motion capture collection to systematically cover a broad range of whole-body motion. Why synchronized object trajectories and meshes make human-object interaction data useful for teaching robots real-world tasks such as carrying, pushing, and pulling. 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. Download this free whitepaper now!