Human Archive builds multimodal robotics dataset infrastructure
Human Archive built multimodal data-collection infrastructure for physical-AI and robotics training.
Evidence notes
- HA-Multi combines vision, stereo depth, tactile gloves, IMUs, and wrist cameras.
- The system targets large-scale robotics and physical-AI dataset generation.
Company context
Human Archive builds multimodal real-world datasets for robotics and world-model development. The company captures human interaction data and processes it through QA, anonymization, and annotation workflows for downstream model training.