Hand Motion Data Library
High-fidelity, ready-to-train hand-motion captures for robotics, teleoperation, and embodied AI. Watch each capture and download the data — FBX and native MREC — right here.

Captured with MANUS MetaGloves Pro — EMF hand tracking, no drift, no occlusion. More capture sources coming to the library.
Click a capture — watch it, grab its data
Pick a motion on the left; the viewer plays it and its .FBX + native .MREC are ready to download.
Why the data is clean: MetaGloves Pro
Millimeter precision
EMF tracking — no drift, no occlusion, easy calibration.
25 D.O.F.
A full anatomical hand model, every joint tracked.
Persistent calibration
Profiles carry across users, takes, and workstations.
Detachable module
One module, many personally-sized sensor gloves.
Align hand data with the whole take
The universal mount attaches third-party trackers — Xsens, OptiTrack, Vive — so hand data lines up with body and scene capture in the same recording. The library grows as you add sources.
Drops straight into your stack
Export to standard file types, stream live, or build a custom pipeline with the MANUS Core SDK.





















Not sure which glove fits?
Compare MetaGloves Pro against the full MANUS VR glove lineup.
Trusted by robotics, research, and studios
After comparing IMU, optical, and vision-based solutions for our robotic foundation model development, MANUS gloves delivered the most reliable and robust performance.
MANUS gloves provide excellent control of dexterous robot hands, delivering precision and data quality that outperform vision-based approaches.
I can say with high confidence, this is by far the best product when it comes to robot hand teleoperation.
Need more than samples?
Knoxlabs delivers custom capture, larger datasets, and full MetaGloves Pro deployments for robotics and embodied-AI teams.