DETECTION OF MULTIPLE OBJECT GROUPS USING A UNIFIED COMPUTER VISION MODEL.

Abstract

The examples provide a multi-object grouping detection model for identifying instances of items arranged in groups of objects of interest using input images of a selected area generated by an image capture device. A grouping detection manager identifies instances of each distinct grouping of items within the image based on the size, shape, and color of the objects, as well as their proximity to one another. Marked image data is generated using the image. This marked image data includes grouping indicators that identify unique groupings within the image. The marked image data is used to determine the number of distinct groupings at a location. The system generates an alert if the detected number of groupings differs from an expected number of groupings for the given location.The system uses grouping detections to update item data, identify out-of-stock items, identify misplaced items, validate automated item-to-location mapping, and verify item IDs.