Abstract: A portable checkout unit automatically generates training data for an automatic checkout system as a customer collects items in a store. A customer uses an item scanner of portable checkout unit to generate a virtual shopping list of items collected in the shopping cart. When the customer adds a new item to the shopping cart or on some regular interval, the portable checkout unit captures images of the items contained by the shopping cart and can generate bounding boxes for each product in each image. The bounding boxes can be associated with item identifiers from previously-generated bounding boxes to identify the items captured by the bounding boxes. Each bounding box paired with an item identifier can then be used as training data for an automated checkout system.

Abstract: An out-of-stock detection system notifies store management that a product is out of stock. The system captures images of a shelf and determines the position product labels thereon. For each product label, a bounding box is generated based on the position of each product label on the shelf. The system then identifies a product for each product label based on information within each product label and, for each product label, stores a product identified for each bounding box. Accordingly, the system performs an out-of-stock detection process that includes capturing additional image data of the shelf periodically that includes each bounding box, providing a portion of the additional image data for each bounding box to a model trained to determine whether the bounding box contains products, sending a notification for a product determined to be out of stock to a store client device based on output from the model.

Abstract: An operating system for a retail store applies AI to detect, from images of store shelves, out of stock and low stock conditions of shelved items based on camera images. The system takes in a set of input states of the store and recommends corrective action to optimize a set of objectives for the store. These objectives could be to optimize Operating Profit. The input states could be current shelf conditions inferred by shelf cameras. The action could be ordering of restocking, changes in future inventory orders, the number of shelf facings per product, price to charge per product, or labor allocations and scheduling for store staff. Through repeated reiterations over an extended period, the system compares actual results with predicted results and retrains itself to minimize the difference and recommend smarter actions over time to “play the game of retail” better and better each day and in each store.

Abstract: A method for generating training examples for a product recognition model is disclosed. The method includes capturing images of a product using an array of cameras. A product identifier for the product is associated with each of the images. A bounding box for the product is identified in each of the images. The bounding boxes are smoothed temporally. A segmentation mask for the product is identified in each bounding box. The segmentation masks are optimized to generate an optimized set of segmentation masks. A machine learning model is trained using the optimized set of segmentation masks to recognize an outline of the product. The machine learning model is run to generate a set of further-optimized segmentation masks. The bounding box and further-optimized segmentation masks from each image are stored in a master training set with its product identifier as a training example to be used to train a product recognition model.

Abstract: A portable checkout unit automatically generates training data for an automatic checkout system as a customer collects items in a store. A customer uses an item scanner of portable checkout unit to generate a virtual shopping list of items collected in the shopping cart. When the customer adds a new item to the shopping cart or on some regular interval, the portable checkout unit captures images of the items contained by the shopping cart and can generate bounding boxes for each product in each image. The bounding boxes can be associated with item identifiers from previously-generated bounding boxes to identify the items captured by the bounding boxes. Each bounding box paired with an item identifier can then be used as training data for an automated checkout system.

Abstract: A retail store automated checkout system uses images, video, or depth data to recognize products being purchased to expedite the checkout process and improve accuracy. All of a store's products, including those not sold in packages, such as fruits and vegetables, are imaged from a series of different angles and in different lighting conditions, to produce a library of images for each product. This library is used in a checkout system that takes images, video, and depth sensor readings as the products pass through the checkout area and remove bottlenecks later in the checkout process. Recognition of product identifiers or attributes such as barcode, QR code or other symbols, as well as OCR of product names, as well as the size and material of the product, can be additional or supplemental devices for identifying products being purchased. In another embodiment, an existing checkout or self-checkout scanner is enhanced with image recognition of products to achieve the same effect.

Abstract: A method for generating training examples for a product recognition model is disclosed. The method includes capturing images of a product using an array of cameras. A product identifier for the product is associated with each of the images. A bounding box for the product is identified in each of the images. The bounding boxes are smoothed temporally. A segmentation mask for the product is identified in each bounding box. The segmentation masks are optimized to generate an optimized set of segmentation masks. A machine learning model is trained using the optimized set of segmentation masks to recognize an outline of the product. The machine learning model is run to generate a set of further-optimized segmentation masks. The bounding box and further-optimized segmentation masks from each image are stored in a master training set with its product identifier as a training example to be used to train a product recognition model.

Abstract: A portable checkout unit automatically generates training data for an automatic checkout system as a customer collects items in a store. A customer uses an item scanner of portable checkout unit to generate a virtual shopping list of items collected in the shopping cart. When the customer adds a new item to the shopping cart or on some regular interval, the portable checkout unit captures images of the items contained by the shopping cart and can generate bounding boxes for each product in each image. The bounding boxes can be associated with item identifiers from previously-generated bounding boxes to identify the items captured by the bounding boxes. Each bounding box paired with an item identifier can then be used as training data for an automated checkout system.

Abstract: An out-of-stock detection system notifies store management that a product is out of stock. The system captures images of a shelf and determines the position product labels thereon. For each product label, a bounding box is generated based on the position of each product label on the shelf. The system then identifies a product for each product label based on information within each product label and, for each product label, stores a product identified for each bounding box. Accordingly, the system performs an out-of-stock detection process that includes capturing additional image data of the shelf periodically that includes each bounding box, providing a portion of the additional image data for each bounding box to a model trained to determine whether the bounding box contains products, sending a notification for a product determined to be out of stock to a store client device based on output from the model.

Abstract: A mobile shopping unit of an automated checkout system includes one or more sensors within a shopping cart that detect when a change to the contents of the shopping cart has occurred. If the mobile shopping unit detects a change, the mobile shopping unit captures image data of the contents of the shopping cart using one or more cameras mounted to the shopping cart. The mobile shopping unit uses the image data to identify the item added to or removed from the cart. The mobile shopping unit applies a machine-learned item identification model to the image data received from the cameras to determine an item identifier for the added or removed item. When the mobile shopping unit determines the identifier for the added or removed item, the mobile shopping unit updates a contents list associated with the customer that stores the items currently collected by the customer.

Abstract: An expedited checkout system allows a customer to generate a virtual shopping list on a client application operating on the customer's mobile device, and expedite checkout from a store using the client application. While shopping within a store, a customer can temporarily secure a mobile device to a portable checkout unit that is attached to a shopping cart. The portable checkout unit can receive item identifiers from an item scanner and transmits item identifiers to the customer's mobile device. The customer's mobile device generates a virtual shopping list of the items collected by the customer and, when the customer is ready to check out of the store, the customer can select a checkout option presented on the mobile device. In response to receiving the checkout selection, the mobile device initiates a checkout transaction to check out the customer from the store.

Abstract: An optical fiber according to Tran U.S. Pat. No. 5,274,728 is improved by chemically etching the surface of the GeO2-based glass perform with an acid solution based on a mixture of HNO3 and water.

Abstract: A method of molding low-temperature glass into a preform for formation by drawing into glass fiber, especially for transmission of mid-IR, involves casting a cladding glass into a mold cavity in the shape of the desired preform to form a cladding layer, and forming a glass core within the cladding layer, wherein the molten cladding glass is drained from the bottom of the mold cavity, forming an annular coating of cladding glass as an annular layer, and the core glass is quickly added within the annular cladding layer to form the glass core with the cladding layer thereabout.

Abstract: High optical quality glass windows, particularly of low melting and low viscosity glasses and substantially free of defects, particularly adapted for high energy laser applications, are made by stirring the molten glass during cooling without the use of a mechanical stirrer within the glass, by rotating the mold or crucible in which the glass is cooling, using a motion which is not entirely a circular and rotary motion.