Methods, systems, and computer-readable media are provided for controlling a detail level of shopping data provided to frictionless shoppers. In one implementation, a method may comprise: receiving image data captured using one or more image sensors in a retail store; analyzing the image data to identify a plurality of product interaction events for at least one shopper in the retail store; determining shopping data associated with the plurality of product interaction events; determining a likelihood that the at least one shopper will be involved in shoplifting; based on the determined likelihood, determining an update rate for updating the at least one shopper with the shopping data; and delivering the shopping data to the at least one shopper at the determined update rate.

A system for determining whether shoppers are eligible for frictionless checkout is disclosed. The system has a processor that obtains image data captured using image sensors positioned in a retail store. The processor analyzes the image data to identify at least one shopper at one or more locations of the retail store. The processor detects, based on the analysis of the image data, at least one product interaction event associated with an action of the at least one shopper at the one or more locations of the retail store. Further, based on the detected at least one product interaction event, the processor determines whether the at least one shopper is eligible for frictionless checkout. In response to a determination that the at least one shopper is ineligible for frictionless checkout, the processor causes delivery of an indicator that the at least one shopper is ineligible for frictionless checkout.

Methods, systems, and computer-readable media are provided for tracking frictionless shopping eligibility relative to individual shopping receptacles. In one implementation a method comprises: obtaining image data captured using a plurality of image sensors; analyzing the image data to identify a shopper at one or more locations; detecting, based on the analysis of the image data, a first product interaction event involving a first shopping receptacle associated with the shopper and a second product interaction event involving a second shopping receptacle associated with the shopper; based on the detected first product interaction event, determining whether the first shopping receptacle is eligible for frictionless checkout; based on the detected second product interaction event, determining whether the second shopping receptacle is eligible for frictionless checkout; and causing delivery of an indicator identifying which of the first shopping receptacle or the second shopping receptacle is ineligible for frictionless checkout.

Methods, systems, and computer-readable media are provided for controlling a detail level of shopping data provided to frictionless shoppers. In one implementation, a method may comprise: receiving image data captured using one or more image sensors in a retail store; analyzing the image data to detect a shopper in the retail store; determining a likelihood that the shopper will be involved in shoplifting; controlling a detail level associated with frictionless shopping data provided to the shopper based on the determined likelihood that the shopper will be involved in shoplifting.

A system for updating virtual shopping carts of shoppers with pay-by-weight products is disclosed. The system has a processor that receive images captured by image sensors. The images depict product interactions between a store associate and shoppers. Each product interaction involves a pay by weight product. The processor analyzes the images to identify the product interactions and associates the pay by weight product involved with each product interaction with a particular shopper among the shoppers. The processor provides a notification to the store associate requesting supplemental information for associating the pay by weight product with the particular shopper. The processor receives the requested supplemental information and uses the analysis of the images and the requested supplemental information to determine the association of the pay by weight product with the particular shopper. The processor also updates a virtual shopping cart of the particular shopper with the pay by weight product.

A method for automatically updating a plurality of virtual shopping carts, includes receiving image data captured in a retail store, wherein a first shopping receptacle and a second shopping receptacle are represented in the received image data, determining that the first shopping receptacle is associated with a first virtual shopping cart and that the second shopping receptacle is associated with a second virtual shopping cart different from the first virtual shopping cart, analyzing the received image data to detect a shopper placing a first product in the first shopping receptacle and to detect the shopper placing a second product in the second shopping receptacle, automatically updating the first virtual shopping cart to include information associated with the first product, and automatically updating the second virtual shopping cart to include information associated with the second product.

A method for identifying products removed from bulk packaging includes receiving one or more images acquired by a camera arranged to capture interactions between a shopper and one or more bulk packages each configured to contain a plurality of products, analyzing the one or more images to identify the shopper and a particular bulk package among the one or more bulk packages with which the identified shopper interacted, receiving an output from at least one sensor configured to monitor changes associated with the particular bulk package, analyzing the output to determine a quantity of products removed from the particular bulk package by the identified shopper, and updating a virtual shopping cart associated with the identified shopper to include the determined quantity of products and an indication of a product type associated with the particular bulk package.

Methods, systems, and computer-readable media are provided for providing a visual indicator indicative of a frictionless checkout status of at least a portion of a retail shelf. In one implementation, a method may comprise: receiving an output from one or more retail store sensors; based on the output from the one or more retail store sensors, determining a frictionless checkout eligibility status associated with the at least a portion of the retail shelf, wherein the frictionless checkout eligibility status is indicative of whether the at least a portion of the retail shelf includes one or more items eligible for frictionless checkout; and causing a display of an automatically generated visual indicator indicating the frictionless checkout eligibility status associated with the at least a portion of the retail shelf.

There is provided a method of image classification. The method includes: providing a set of category mapping discriminators, each corresponding to a respective category, wherein each category mapping discriminator of the set of category mapping discriminators is configured for discriminating features relating to input images that belong to the respective category of the category mapping discriminator; extracting a plurality of features from an input image using a machine learning model; determining, for each of the set of category mapping discriminators, an output value based on the plurality of extracted features using the category mapping discriminator; and determining a classification of the input image based on the output values of the set of category mapping discriminators.

One embodiment of the present invention sets forth a technique for performing spatial propagation. The technique includes generating a first directed acyclic graph (DAG) by connecting spatially adjacent points included in a set of unstructured points via directed edges along a first direction. The technique also includes applying a first set of neural network layers to one or more images associated with the set of unstructured points to generate (i) a set of features for the set of unstructured points and (ii) a set of pairwise affinities between the spatially adjacent points connected by the directed edges. The technique further includes generating a set of labels for the set of unstructured points by propagating the set of features across the first DAG based on the set of pairwise affinities.