PoXiao Intelligence and Tinno Mobile Launch SeePhys Hub to Advance Full-Body, Full-Modal Human Data Collection

SeePhys Hub is a full-body, full-modal data collection system designed for embodied intelligence. It consists of a four-camera head-mounted ring, two wristbands, three waist and leg motion trackers, and a pair of haptic gloves, capable of synchronously capturing first-person 360° panoramic vision, full-modal hand data (pose tracking and tactile feedback), and full-body motion information.

Robot data collection is currently undergoing a lightweight evolution from teleoperation to UMI to Ego, with devices becoming lighter and collection procedures becoming simpler. However, this simplification also leads to the loss of critical information—Ego can record hand movements but struggles to fully capture full-body motion and cannot sense pressure changes when hands contact objects, yet these simplified interaction details are often essential for robots to complete tasks.

SeePhys Hub seeks to find a balance between lightweight design and data completeness, using multi-device collaborative collection to compensate for the perceptual blind spots of single lightweight devices. In the field of embodied intelligence, high-quality multimodal real-world interaction data has always been a key bottleneck in training robot manipulation models, and the improvement of such data collection infrastructure helps lower the barrier to data acquisition.

Embodied intelligence is a major frontier in the current AI landscape, and one of its core challenges is enabling models to acquire sufficient real-world interaction experience. The emergence of full-body, full-modal data collection systems reflects the industry's rising demands for data quality and signals that the infrastructure for embodied intelligence training data is accelerating toward maturity.

This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.