IP Library Granted Patent US 12,536,548
Granted Patent B2
US 12,536,548 · App. 16/368,372 · Granted Jan 27, 2026

Frictionless and unassisted return processing

Inventors: Chario Bardoquillo Maxilom (Ibabao Cordova, PH); Ferdinand Salarda Acedera (Cebu, PH); John Wilte Go Mabute (Umapad Mandaue, PH)
Assignee: NCR Voyix Corporation
G06Q20/407G06Q20/40145G06Q30/016
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,536,548
App. No.
16/368,372
Granted
Jan 27, 2026
Kind
B2
Abstract

Images capture an object that a person carries into a frictionless store as a potential frictionless return item. When the potential frictionless return item is placed on a shelf or a countertop in a designated area, the potential frictionless return stem is identified as a returned stem A transaction history linked to an account of the person is retrieved and the returned item matched to a previously purchased item in the transaction history. Assuming return conditions are met, the account is automatically credited an item price associated with the previously purchased item and a user interface is automatically initiated on a registered mobile device of the person providing a receipt for the refund and the return. The person may then leave the store or continue shopping in the frictionless store and the person and any items taken by the person are tracked for frictionless purchase.

Claims (51)

1 . A method, comprising:

executing, by a processor, executable instructions from a non-transitory computer-readable storage medium causing the processor to perform operations comprising:

obtaining a stream of images from a plurality of cameras;

maintaining a bounding box for the stream of images to track a person while the person is inside a store using camera metadata associated with each camera;

determining, based on an area of the store designated for returns and refunds, a particular camera where a clear image is captured of the designated area, the particular camera having associated metadata and identifier associated with images captured of the designated area;

tracking the person inside the store and confirming possession of an item by the person by monitoring the bounding box for the stream of images, wherein tracking comprises monitoring the person upon entry into the store until the item is placed in the designated area;

determining, by monitoring the bounding box, that the person has placed the item in the designated area of the store;

determining a particular image captured of the designated area that includes a view of the item;

cropping out, from the particular image, background pixels not associated with the item to produce a modified item image for the item;

obtaining attributes of the item associated with the particular image using the particular camera metadata;

adding additional attributes of the item to the particular camera metadata;

identifying the item, based on the modified item image, the attributes, and the additional attributes;

assigning an identity to the person based on a certain image captured of the person or based on an event;

matching the identity of the person to an account associated with a consumer of the store;

obtaining a transaction history for the consumer based on the account;

matching the item to a previously purchased item in the transaction history associated with the consumer;

crediting the account of the consumer with an item price associated with the previously purchased item as a refund for the item, without any action required of the consumer other than leaving the item in the designated area; and

providing a receipt for the refund to a consumer-operated device, wherein the receipt is displayed in real time to the consumer on the consumer-operated device.

2 . The method of claim 1 , wherein matching the identity of the person comprises associating a check-in event with the particular image, identifying the consumer, and linking the account to the consumer.

3 . The method of claim 2 , wherein associating the check-in event comprises detecting the check-in event through one or more of: facial recognition of the consumer, a mobile location reported that indicates the consumer has entered the store, or a scan of a bar code performed by the consumer-operated device upon entry into the store.

4 . The method of claim 1 , wherein matching the item to a previously purchased item comprises confirming a refund policy associated with the store or the item, based on the transaction history.

5 . The method of claim 4 , wherein crediting comprises sending an application message to a mobile application of the consumer-operated device after determining that the item cannot be returned based on a failure to comply with the refund policy.

6 . The method of claim 1 , wherein crediting comprises causing initiating of a mobile application on the consumer-operated device.

7 . The method of claim 1 , wherein crediting comprises requesting a refund type from a mobile application that processes on the consumer-operated device.

8 . The method of claim 7 , wherein requesting comprises removing a credit made to the account and processing the refund type based on receiving a response provided by the consumer through the mobile application and determining the refund type identified in the response does not correspond with the crediting of the account.

9 . A system, comprising:

at least one processor;

a non-transitory computer-readable storage medium storing instructions, that when executed by the at least one processor cause the at least one processor to perform operations, comprising:

obtaining a stream of images from a plurality of cameras;

maintaining a bounding box for the stream of images to track a person while the person is inside a store using camera metadata associated with each camera;

determining, based on an area of the store designated for returns and refunds, a particular camera where a clear image is captured of the designated area, the particular camera having associated metadata and identifier associated with images captured of the designated area;

tracking the person inside the store and confirming possession of an item by the person by monitoring the bounding box for the stream of images, wherein tracking comprises monitoring the person upon entry into the store until the item is placed in the designated area;

determining, by monitoring the bounding box, that the person has placed the item in the designated area of the store;

determining a particular image captured of the designated area that includes a view of the item;

cropping out, from the particular image, background pixels not associated with the item to produce a modified item image for the item;

obtaining attributes of the item associated with the particular image using the particular camera metadata;

adding additional attributes of the item to the particular camera metadata;

identifying the item, based on the modified item image, the attributes, and the additional attributes;

assigning an identity to the person based on a certain image captured of the person or based on an event;

matching the identity of the person to an account associated with a consumer of the store;

obtaining a transaction history for the consumer based on the account;

matching the item to a previously purchased item in the transaction history associated with the consumer;

crediting the account of the consumer with an item price associated with the previously purchased item as a refund for the item, without any action required of the consumer other than leaving the item in the designated area; and

providing a receipt for the refund to a consumer-operated device, wherein the receipt is displayed in real time to the consumer on the consumer-operated device.

10 . The system of claim 9 , wherein matching the identity of the person comprises associating a check-in event with the particular image, identifying the consumer, and linking the account to the consumer.

11 . The system of claim 10 , wherein associating the check-in event comprises detecting the check-in event through one or more of: facial recognition of the consumer, a mobile location reported that indicates the consumer has entered the store, or a scan of a bar code performed by the consumer-operated device upon entry into the store.

12 . The system of claim 9 , wherein matching the item to a previously purchased item comprises confirming a refund policy associated with the store or the item, based on the transaction history.

13 . The system of claim 12 , wherein crediting comprises sending an application message to a mobile application of the consumer-operated device after determining that the item cannot be returned based on a failure to comply with the refund policy.

14 . The system of claim 9 , wherein crediting comprises causing initiating of a mobile application on the consumer-operated device.

15 . The system of claim 9 , wherein crediting comprises requesting a refund type from a mobile application that processes on the consumer-operated device.

16 . The system of claim 15 , wherein requesting comprises removing a credit made to the account and processing the refund type based on receiving a response provided by the consumer through the mobile application and determining the refund type identified in the response does not correspond with the crediting of the account.

Assignments (2)
CHANGE OF NAME Recorded Nov 9, 2023
From: NCR CORPORATION
To: NCR VOYIX CORPORATION
Reel/Frame 065532/0893 →
SECURITY INTEREST Recorded Oct 25, 2023
From: NCR VOYIX CORPORATION
To: BANK OF AMERICA, N.A., AS ADMINISTRATIVE AGENT
Reel/Frame 065346/0168 →