Access to
100k+ SKUs
Full control over the data
Award-winning RPA Partner
Expert Support
​
STEP 1
Select SKUs of interest
With direct access to 100,000+​ SKUs on demand and 2,000+ of the most popular SKUs out-of-the-box, we have you covered.
Following best-in-class industry standards, SKUs are organised by category. Search through the catalogue by product name or EAN.
​
If you can't find your SKU in the catalogue, simply upload 6 photos and we'll generate a digital twin of the SKU in less than 48h. Finally, we also offer you the option to upload your own 3D assets.
1
Pick a category
Start by choosing a category of objects. Check our public catalogue and leverage all existing categories.
2
Create your catalogue
Next, define your private catalogue by selecting specific objects from a category.
3
Import your data
Still missing an SKU? Import 6 photos of each side of the product and we'll have digitise it for you. If you have the 3D asset, you can simply upload it.
STEP 2
Synthetic Dataset
It only takes 3 clicks to generate a highly diverse dataset. Choose from a large variety of pre-built scene configurations that have been industry validated. There are many out-of-the-box configurations available, such as SKU detection for shelves, smart carts or conveyor belts.
Easily customise each dataset using no-code mode or use expert mode to control low-level specifics of the scene, such as camera type, camera angle, scene lighting or occlusion rate.
1
Define SKU catalogue
Start by choosing a category of objects. Check our public catalogue and leverage all existing categories.
2
Select a scene
Whether your use-case is shelf monitoring, smart cart or cashierless checkout, out-of-the-box configurations are waiting for you.
3
Customise the dataset
Have deeper control over the dataset by specifying key variables such as lighting type, camera positioning, occlusion rate, and much more.
STEP 3
Validate and export your data
Once the dataset is ready, you will find it under your Datasets collection. ​
For your freshly generated dataset, a high-level summary is available. Next, visualise the generated images, with or without annotations and labels. ​
Finally, export your dataset locally in the standard COCO format.
1
Validate meta-data
Take a high-level peak into your dataset to check that the output meets your expectations.
2
Visualise your data
Visually check samples from the dataset to validate their quality and relevance for your specific use-case.
3
Download the data
Export the data locally. You'll have instant access to the images and the annotations files in the industry-wide COCO standard.
