Target Brands, Inc.

Ethan Sommer

Christopher Mark Brakob

Christopher M Brakob

Prakash Mall

Dharmavaram Arbaaz

Neha Dixit

Arun Patil

Arun Vaishnav

The disclosed technology provides for detecting a scam attempt in a physical retail environment. A system can include a camera that captures image data of a checkout area, a cash register that generates event data during a checkout process, and an edge computing device that accesses scam detection criteria identifying combinations of visual and event features corresponding to checkout scams, receive the image data, receive the event data, which includes (i) an indication that the checkout process is a cash transaction and/or (ii) a cash register drawer open event, detect, based on processing the image data, physical movement of a form of payment in or around the checkout area, correlate timestamps for the physical movement with the event data, generate an indication of the checkout scams based on satisfying the scam detection criteria, and return the indication of the checkout scams being performed.

Computer-vision based detection of scams at physical point of sale terminals

The disclosed technology provides for generating best images of a person in a retail environment. A method may include receiving, by an edge computing device from a camera, a continuous stream of image data of the retail environment, detecting, using object detection techniques, a person in the image data, the image data including a group of images that are part of a time series, generating bounding boxes for each of the group of images around the person based on detecting the person as they move in the images, identifying, based on applying a features model to each bounding box, at least one feature of the group of images depicting the person, selecting a subset of the bounding boxes having at least one feature that satisfies best images criteria, the subset having best images of the person, and returning the best images of the person.

Automatically generating best digital images of a person in a physical environment

Disclosed are systems and methods for determining whether an unknown product matches a scanned barcode during checkout. The system includes a checkout lane having a flatbed scanning area with scanning devices and a point of sale (POS) terminal that scans a product identifier of an unknown product, identifies a product associated with the scanned product identifier, and transmits, to a computing system, product information. An overhead camera idnentifies, based on detecting an optical signal from the POS terminal, that a scanning event occurred, captures image data of the unknown product, and transmits, to the computing system, the image data. The computing system generates machine learning product identification models for identifying unknown products, identifies candidate product identifications for the unknown product based on applying the models to the image data, and determines, based on the candidate product identifications and the information about the product, whether the unknown product matches the product.

Identifying barcode-to-product mismatches using point of sale devices and overhead cameras

The disclosed technology provides for automatically detecting and responding to potentially suspicious or risky activity in a retail environment. A method can include receiving, from monitoring devices in a retail environment, a stream of activity data, applying a model to the stream of activity data to identify a portion of the data corresponding to guest activity during a checkout process, identifying whether a risk event is associated with the activity, determining a guest risk impact score, selecting (i) a particular manual response from among candidate manual responses and (ii) a particular automated response from among candidate automated responses based on the risk impact score satisfying manual response criteria and/or automated response criteria, transmitting instructions to a POS terminal to implement the particular automated response, and/or transmitting instructions to implement the particular manual response to one or more mobile devices, that prompt employees to perform the manual response.

Risk-based adaptive responses to user activity in a retail environment

Described herein are systems and methods for determining whether a scanning motion occurred during a checkout process. The system includes a checkout lane having a scanning area that receives products to be purchased by a user, scanning devices, a point of sale (POS) terminal that identifies a product based on a scan, using the scanning devices, of a product identifier for the product as the product is moved through the scanning area, and an overhead camera that captures image data of the user's body movements and transmits, to a computing system, the image data. The computing system can, during runtime, identify whether a scanning motion occurred during the user's body movements based on application of one or more motion identification models to the image data and determine, based on identification of the scanning motion, that the user performed an affirmative scan during the checkout process.

Identifying scanning motions during checkout using overhead cameras

Described herein are systems and methods for generating a watch list of users who pose specific security threats to a store. The method can include retrieving, by a computer system from a data store, case files that document activity that poses a security threat by a user at the store, predicting, based on applying prediction models to the case files, future activity associated with the case files, determining threat scores for the case files based on the predicted future activity, ranking the case files into a candidate list from highest to lowest threat score, generating a watch list for the store that includes a subset of the ranked case files based on which case files pose a greatest current threat to the store, generating summary videos for each case file in the watch list, and transmitting the watch list and summary videos to a user device.

Generating watch lists for retail stores based on unstructured data and system-based inferences

Described herein are systems and methods for generating security event case files with unstructured data. For example, the method can include receiving, by a computing system, unstructured data and system-based inferences from devices positioned throughout a store, and adding structure to the unstructured data and system-based inferences based on applying one or more structuring models. Adding structure can include labeling the data and system-based inferences, classifying them into security event categories, and identifying objective identifiers to identify users in the data and system-based inferences. The method also can include generating case files for each of the objective identifiers, where the case files include the associated data. The method can include determining whether the case files satisfy alerting rules. The case files can then be reported out and acted upon (e.g., based on satisfying the alerting rules) and/or stored for subsequent analysis and use.

Generating security event case files from disparate unstructured data

A computer-implemented method includes determining that a group of items left a space together based on RFID exit times and determining that at least one item of the group of items left the space without authorization by evaluating transactions that took place within a first time period before the at least one item left the space. A single investigation file is automatically created that includes the group of items that left the space together.

Item tracking system

Disclosed herein are systems and methods for determining whether an unknown product matches a scanned barcode during a checkout process. An edge computing device or other computer system can receive, from an overhead camera at a checkout lane, image data of an unknown product that is placed on a flatbed scanning area, identify candidate product identifications for the unknown product based on applying a classification model and/or product identification models to the image data, and determine based on the candidate product identifications, whether the unknown product matches a product associated with a barcode that is scanned at a POS terminal in the checkout lane. The classification model can be used to determine n-dimensional space feature values for the unknown product and determine which product the unknown product likely matches. The product identification models can be used to determine whether the unknown product is one of the products that are modeled.

Growing community of inventors

Minneapolis, MN, United States of America

Ethan Sommer