Overlap detection for an item recognition system

A non-transitory, computer-readable medium storing instructions that, when executed by a processor, cause the processor to:

• access a top image comprising an image captured by a top camera of an item recognition system, wherein the top image depicts a first item and a second item on a receiving surface of the item recognition system;
• access one or more peripheral images, each comprising an image captured by a peripheral camera of one or more peripheral cameras of the item recognition system;
• generate a pixel-wise mask for the top image based on the top image, wherein pixel-wise mask indicates one or more portions of the top image where an item is depicted;
• apply an overlap detection model to the top image, the one or more peripheral images, and the pixel-wise mask to detect whether the first item overlaps with the second item, wherein the overlap detection model is a machine-learning model trained to detect overlapping items in top images based on the top images, peripheral images, and pixel-wise masks of the top images;
• and
• responsive to detecting that the first item overlaps with the second item, present a notification of the overlap through a user interface of the item recognition system.

A method comprising:

• accessing a top image comprising an image captured by a top camera of an item recognition system, wherein the top image depicts a first item and a second item on a receiving surface of the item recognition system;
• accessing one or more peripheral images, each comprising an image captured by a peripheral camera of one or more peripheral cameras of the item recognition system;
• generating a pixel-wise mask for the top image based on the top image, wherein pixel-wise mask indicates one or more portions of the top image where an item is depicted;
• applying an overlap detection model to the top image, the one or more peripheral images, and the pixel-wise mask to detect whether the first item overlaps with the second item, wherein the overlap detection model is a machine-learning model trained to detect overlapping items in top images based on the top images, peripheral images, and pixel-wise masks of the top images;
• and
• responsive to detecting that the first item overlaps with the second item, presenting a notification of the overlap through a user interface of the item recognition system.