Abstract: An imaging system capture images of products on shelves in a minimally invasive way. The images can be processed to assess restocking needs and other information. The imaging system includes multiple shelf-mountable imaging devices that can be positioned throughout one or more retail locations. Certain types of imaging devices may be self-contained units having replaceable or rechargeable internal power sources and/or wireless communication interfaces. Certain types of imaging device are configured to take images only when the field of view is not blocked by traffic (e.g., consumers, employees, etc.). Certain types of imaging devices are low-profile and/or are camouflageable to enhance the user experience.

Abstract: A disclosed server system for dynamically generating and serving content for presentation across different channels can perform the following operations: receive a request from a client device for a subset of content elements, the request designating a desired channel for presenting content in a page, identify the subset having at least one content attribute that satisfies one or more subset creation criteria indicating a category identifier for a slot in a page template used for generating the page, dynamically generate the subset by: for each content element in the subset, dynamically adjusting presentation of the singular instance of the content element based on presentation requirements of the desired channel, the singular instance being commonly used across all device interface types, and serve the subset of the content elements to the client device, which includes the singular instances of the content elements dynamically adjusted for presentation in the desired channel.

Abstract: Systems and methods for monitoring webpage traffic. A web server can generate a JavaScript wrapper and wrap requested webpage code in the JavaScript wrapper. A client device in communication with the web server can receive a user request to load a webpage in a web browser, receive, from the web server, webpage code for the requested webpage that is wrapped in the JavaScript wrapper and includes an allowlist of allowed domains, execute the wrapped webpage code in the web browser, receive a domain request, identify the requested domain from the allowlist, execute the request if the domain is in the allowlist, and transmit, to the web server, run-time information associated with the domain request. The web server can determine, based on the run-time information, a frequency that the domain request was made, add the requested domain to a proposed allowlist, and generate a proposed modification to the JavaScript wrapper.

Abstract: Methods and systems are described for providing hierarchical demand forecasting for state space reduction. Using a hierarchical architecture, a base model may be trained to capture a range of shared structure in a first data set that can be used to draw inferences on using smaller sets of data representative of the “whole picture.” For example, training data may be sampled and prepared and used to train a base model in a first stage. In a next stage, one or more downstream models may be trained on the structure and samples of uncensored demand generated by the base model to produce forecasts for items and locations, including items and locations for which there may be little or no historical data. The downstream models that would otherwise require a large amount of data for training can be generated on demand using less training data, training time, computing processing, and memory.

Abstract: 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.

Abstract: In some implementations, a method performed by data processing apparatuses includes receiving, by an item verification engine, a scan notification indicating that a mobile computing device has scanned an item. In response to receiving the scan notification, the item verification engine provides an instruction for a cart computing system to activate item sensors on a shopping cart. The item verification engine receives sensor data collected by the item sensors as a result of the item having been placed in the shopping cart. The item verification engine receives item verification data for verifying the scanned item from the mobile computing device, performs a verification of the item based on the sensor data and the item verification data, and provides verification results for presentation by the mobile computing device.

Abstract: Computer-implemented methods and systems to present a graphical user interface in a virtual reality retail environment of a retailer online service are provided. The methods and systems allow one or more guests to navigate products/services provided by the retailer online service within virtual reality retailer environments in an easy, accessible, and entertaining manner. A card ecosystem guest interface is provided that allows guests to search, filter and examine different product/services and can display one or more three-dimensional (3D) cards that provide information regarding the products/services available to purchase from the retailer online service. The guest can select a 3D card to view additional information regarding the product/service and to provide a miniature 3D model of the product/service. The guest can also generate a scaled 3D model of the product/service to interact with within the virtual reality retail environment.

Abstract: Methods and systems for securely transferring and merging an electronic shopping cart curated by an affiliate online service for purchase on a retailer online service is disclosed. The retailer online service provides an anonymous token to the affiliate online service that allows the affiliate online service to curate an anonymous electronic shopping cart for a user that does not include any identifying information that links the user to a guest account of the retailer online service. When the user is redirected from the affiliate online service to the retailer online service, the anonymous token is validated. Upon validation of the anonymous token, the retailer online service obtains identifying information linking the user to the guest account of the retailer online service and obtain a pre-existing electronic shopping cart linked to the guest account. The retailer online service then merges the anonymous electronic shopping cart with the pre-existing electronic shopping cart.

Abstract: The disclosed application relates to a tool by which a user may create a cloud workspace that includes a data memory space, as well as a tool for automatically identifying ad-hoc analyses on that data. The solution allows a user to connect to data sources using SQL or GUI tools, combine data from different data sources, prepare and clean the data, mine the data for insights, and move that data into downstream reporting tools for visualization. The system is linked to a code repository to allow data scientists to execute code from the code repository in trial data spaces, investigate that data, and prepare more in-depth analytics for downstream reporting tools.

Abstract: A platform, web application, system, and methods are disclosed for creating digital 3D scenes having digital 3D objects and for creating images of the digital 3D scenes. The platform can be communicatively coupled with the web application. The platform can apply a multi-stage method to convert high-fidelity digital 3D objects into low-fidelity digital 3D objects, store a mapping between them, and transmit low-fidelity digital 3D objects to the web application. The web application can include user interfaces for manipulating low-fidelity digital 3D objects to create low-fidelity digital 3D scenes. The platform can automatically convert low-fidelity digital 3D scenes received from the web application into high-fidelity digital 3D scenes and create high-fidelity 2D images.