Show prices in USD INR
Start a project Customer sign in
Computer Vision

Retail Shelf & Out-of-Stock Vision Set

42K shelf photographs with 1.3M product boxes, out-of-stock gaps and planogram compliance labels.

42K images · 1.3M boxesStatic release (v2026.1)JPG + COCO & YOLO labels
Records
42K images · 1.3M boxes
Size
≈ 24 GB
Formats
JPG + COCO & YOLO labels
Coverage
6 countries · convenience, supermarket, pharmacy

Shelf images captured in convenience stores, supermarkets and pharmacies across six countries, annotated for three tasks at once: product detection (1.3M boxes over 900 classes), empty-gap segmentation for out-of-stock detection, and a planogram compliance flag per shelf section.

Lighting, angle, reflections and occlusion are deliberately varied and labeled, so you can measure where a model degrades instead of discovering it in the field.

Schema

FieldTypeDescription
image_id / file_name string Identifier and path in the archive
width / height integer Image dimensions in pixels
country / store_format string Where the photograph was taken
boxes array class_id, class_name and bbox per product
gap_regions array Polygons marking empty shelf space
planogram_compliant boolean Section matches the reference layout
lighting / occlusion_level string Capture conditions
split string train, validation or test

Sample rows

sample.jsonl
{
    "image_id": "SHELF-020931",
    "file_name": "images/SHELF-020931.jpg",
    "width": 3024,
    "height": 4032,
    "country": "ES",
    "store_format": "supermarket",
    "lighting": "mixed",
    "occlusion_level": "medium",
    "planogram_compliant": false,
    "split": "train",
    "boxes": [
        {
            "class_id": 117,
            "class_name": "biscuits_200g",
            "bbox": [
                412,
                1880,
                655,
                2210
            ]
        },
        {
            "class_id": 342,
            "class_name": "detergent_1kg",
            "bbox": [
                1510,
                960,
                1902,
                1498
            ]
        }
    ],
    "gap_regions": [
        [
            [
                700,
                1900
            ],
            [
                980,
                1900
            ],
            [
                980,
                2180
            ],
            [
                700,
                2180
            ]
        ]
    ]
}

Illustrative sample showing the structure of the data. The full dataset is delivered after purchase.

Popular use cases

  • Out-of-stock detection from shelf images
  • Planogram compliance scoring
  • Product recognition models
  • Shelf-share and facings measurement

Delivery & licence

Delivery
Delivered within 24 hours via a secure download link sent to your email and customer account.
Licence
Internal business use licence — see Terms and Conditions
Refunds
Covered by our Refund & Cancellation Policy
Related

You may also like

All datasets
Computer Vision

Invoice & Receipt Document AI — 9 Tax Regimes

30K synthetic invoices, receipts and bills from nine tax regimes with field-level bounding boxes.

Records
30K documents + annotations
Refresh
Static release (v2026.1)
PNG & PDF + COCO… CSV
$249 one-time
E-commerce & Retail

Quick-Commerce Stock & ETA Index

Half-hourly stock, delivery-time and price snapshots from 900 dark stores in 14 cities.

Records
58M store × SKU snapshots
Refresh
Every 30 minutes · monthl…
Parquet CSV
$349 one-time
E-commerce & Retail

Consumer Electronics Price & Demand Panel

24 months of matched prices, sales rank and review velocity for 120K SKUs across six marketplaces.

Records
120K SKUs · 24 months daily
Refresh
Daily observations · mont…
Parquet CSV
$249 one-time
Let’s talk data

Have a dataset or model in mind?

Tell us what you are trying to predict, automate or understand. We will reply within one business day with an approach, a timeline and a quote.