Ncr Corporation

Ncr Voyix Corporation

Stefan Bjelcevic

Brent Vance Zucker

Adam Justin Lieberman

Christian Lee McDaniel

Matthew Robert Burris

Jodessiah Sumpter

Georgiy Pyantkovs'ky

Justin Paul

Multiple images of multiple items are captured of a transaction area during a checkout. The Red-Green-Blue (RGB) data associated with each item image patch is collected across the images and provided as input by a Machine-Learning Model (MLM), which returns an item code for the item. When the given MLM is unable to satisfactorily predict an item code for a given set of image patches, the patches associated with the images are presented to an operator of a checkout and the operator is asked to scan an item barcode for that item. The patches are labeled within the images with the item code and additional images of the item are captured and labeled with the item code when the barcode is scanned by the operator. The labeled images are used in a subsequent training session with the MLM to improve its item recognition accuracy for the item.

Incremental training of a machine-learning model (MLM) for multiview item recognition

Multiple images of a designated area are taken. The designated area comprises multiple items that are to be identified from the images. Depth information and Red, Green, Blue (RGB) data from each image is processed to create a point cloud for each image of the designated area. The point clouds are patched together or synchronized into a single point cloud for the designated area. Known background pixels associated with backgrounds for each image are removed from the single point cloud. The depth information and RGB data for the single point cloud is clustered together and bounding boxes are placed around each item in the single point cloud. At least the RGB data for each bounding box is provided to a machine-learning model (MLM) and the MLM returns an item code for the corresponding item. The item codes are fed to a transaction manager for a transaction associated with a customer.

Multi-item product recognition for checkouts

Locations of a customer's device are tracked relative to an order placed with an establishment. The establishment is instructed to begin order preparation to coincide with the customer's arrival for the order. Customer's arrival at a pickup location is detected within a geofenced area for a premises of the establishment's location. Sensor data provided for the geofenced area is processed to track a vehicle of the customer within the geofenced area and to identify a specific location within the geofenced area where the vehicle is parked. Delivery staff are notified that the customer has arrived for the order and directed to the specific location where the vehicle is parked.

Intelligent order pickup processing

A system can visually track which items in a store are selected for purchase by a shopper. The system can form a virtual shopping cart by analyzing multiple images, over time, to determine which purchasable items are located with the shopper, such as in a physical shopping cart, in a basket, or held by the shopper. By analyzing multiple images, over time, the system can account for items misidentified in one or more images, or fully or partially obscured in one or more images as the shopper traverses the store. Alternatively, the system can form a virtual shopping cart by analyzing instances in which a purchasable item is removed from a shelf or placed on a shelf. Items removed from, but not returned to, a shelf can be considered to be selected for purchase. The system can include a frictionless checkout that charges the shopper for the selected items.

Constructing shopper carts using video surveillance

Various embodiments herein each include at least one of systems methods, software, and data structures for electronic lock control and audit. The various embodiments, although widely applicable, are well suited for retail use where higher-value and access-controlled merchandise is stored in locked cages, closets, rooms, and cabinets to prevent theft. One method embodiment includes receiving, via a network from a mobile device app, an unlock request including an employee identifier and a lock identifier to unlock a secured resource of the lock identifier. The method proceeds in processing the request by validating the employee identifier is authorized to unlock the lock of the lock identifier at a current date and time and retrieving a key-code that will unlock the lock of the lock identifier. The method then proceeds with transmitting the key-code to the mobile device app to enable unlocking of the lock.

Growing community of inventors

Roswell, GA, United States of America

Stefan Bjelcevic