7,000 hours · 2 packages
Retail & Commercial Services robot datasets
First-person video of real retail and hospitality work: restocking shelves, picking and placing products, organizing service items, setting tables and preparing guest rooms.
Packages
Retail & Commercial Services packages
| Package | Representative tasks | Paradigm | Modalities | Hours | |
|---|---|---|---|---|---|
| Retail Egocentric Task Processes | shelf restocking, product pick & place, service item organizing | Egocentric video | Video, Labels & task metadata | 5,000 | Request sample |
| Hospitality Service Operations | guest-room amenity organizing, table setting, service item delivery | Egocentric video | Video, Labels & task metadata | 2,000 | Request sample |
Hours are estimates as of October 2026. Detailed statistics (episodes, environments, object categories and more) available under NDA.
What teams train with this data
- Service-robot teams pre-training task understanding for store and hotel environments.
- VLA teams learning long-horizon task structure from step-by-step egocentric records.
- Researchers building world models of commercial service workflows.
Delivery formats
Delivered in your training format: LeRobot, RLDS, HDF5, MCAP / ROS bag, or custom. Every package passes automated alignment checks and multi-stage human review, and is de-identified before delivery.
Request a Retail & Commercial Services sample.
Tell us what you need. We reply within 3 business days.