Vision-based frictionless self-checkouts for small baskets

A method comprising:

• obtaining three images within a scan zone during a checkout transaction at a terminal, the three images comprising a first image depicting a first overhead view of the scan zone at a first angle, a second image depicting a second overhead view of the scan zone at a second angle different from the first angle and a third image depicting a third overhead view of the scan zone at a third angle different from the first angle and the second angle;
• processing the three images, wherein processing the three images comprises calculating, for the first image a respective distance between each item in the scan zone and a first camera that captured the first image;
• determining an item location for each item present within the scan zone from the three images based on the processing;
• and
• providing, to the terminal, a respective item identifier for each item in the scan zone to enable item pricing to be added to a transaction record associated with the checkout transaction.