12,000+ episodes | SKU visual recognition, shelf state changes, and retail robotics operational decision-making
| Dataset Name | Retail Operations and Restocking Dataset |
| Total Data Volume | 12,000+ episodes |
| Data Type | Videos, SKU images, bar/2D codes, price tag OCRs, inventory records |
| Video Specifications | 1920×1080 @ 30 FPS |
| Delivery Format | LeRobot 3.0 / RLDS / HDF5 / JSON Metadata |
| License Type | Licensed Data or Custom Projects |
Each piece of data contains complete structured fields, covering all requirements from basic metadata to professional analysis.
| Field Category | Typical Fields |
|---|---|
| Shelf and Merchandise Video | Video, shelf level, perspective trajectory, lighting conditions, door area, camera height |
| SKU Identification Information | Product barcode, SKU ID, name, specifications, brand, packaging image, category, shelf life |
| Replenishment Operation Annotations | Pick up goods, place, adjust direction, mark out-of-stock, expired, frame-to-frame, action time |
| Inventory and Price Tags | Changes in inventory quantity, price tags, promotion labels, ranking statistics, member/original prices |
| Metadata and Compliance | Store code, shelf code, collection time frame, authorization scope, privacy de-identification annotations |
The following is a de-identified structured JSON sample illustrating the typical data format.
{ "episode_id": "RTL-2026-0023411", "store": "supermarket_east_02", "shelf_id": "B12-3", "video": {"rgb": "rgb_0023411.mp4", "depth": "depth_0023411.mkv", "fps": 30}, "sku_items": [ {"sku_id": "SKU-88421", "name": "sample 250ml", "barcode": "6901234567890", "brand": "sample value"} ], "actions": [ {"time": 5.2, "action": "pick", "sku": "SKU-88421", "source": "cart"}, {"time": 8.7, "action": "place", "sku": "SKU-88421", "target": "B12-3-2"} ], "before_after": {"out_of_stock": ["B12-3-1"], "restocked": ["B12-3-2"]}, "inventory_delta": {"SKU-88421": +6}
}The robots are trained to independently identify missing goods on supermarket shelves and complete replenishment.
Multiple cameras integrated with inventory data to train smart convenience store operations.
Automatically detects compliance with product display, price tags, ranking, and promotions.
Large volumes of commercial images and barcode data are used to train retail-scene object detection and recognition models.
Pre- and post-restocking data changes are used to train inventory alert and demand forecasting models.
Shelf-area video supports pick-and-place behavior recognition and heat-map analysis.
Changsha Langhui Information Technology Co., Ltd. The DataAssetsAPI platform is dedicated to providing high-quality, compliant data assets for AI companies and research institutions. It supports flexible combination by dimension and by quantity scale.
Data Sample Preview · Customized Data Solutions · Professional Technical Support