DigestAI news desk

Cut through the AI noise.

Robotics & Physical AI5 min read

Innodata opens motion-capture lab for humanoid robot training data

Innodata Inc. announced the launch of a motion-capture lab in Ridgefield Park, New Jersey, designed to generate high-quality training data for humanoid robots and other physical AI systems. The facility uses Vicon’s high-precision infrared cameras to capture 3D motion data directly from bodies—human or robotic—with sub-millimeter accuracy, addressing a critical bottleneck in robotics…

1 source

Key points

  • Innodata’s lab uses Vicon’s sub-millimeter cameras to capture 3D motion data for humanoid robots and physical AI
  • Facility offers data packages and custom projects, plus independent validation of robot performance and safety
  • Company claims 80% of egocentric data is often unusable, pushing for smarter data collection methods

The lab offers end-to-end services, from data collection to model evaluation, aiming to compress development cycles and improve robot safety. Customers can purchase off-the-shelf data packages or commission custom projects, while Innodata also validates robot performance against customer specifications. The company emphasizes the need for real-world data, noting that simulations alone cannot replicate the complexity of human-robot interactions. Tanner highlighted challenges like incomplete teleoperation data and the inefficiency of collecting vast amounts of low-quality data, suggesting a focus on refining data collection methods to maximize utility.

Full story from The Robot Report · by Eugene DemaitreOpen source ↗

Innodata opens motion-capture lab to help humanoids move more like people

The Robot Report · 30 September 2026

Loading the full article…

This text was published by The Robot Report and written by Eugene Demaitre. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page
Innodata Inc.ViconRahul SinghalFranklin TannerAndrew Knox

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Robotics & Physical AI

All →

Related stories