Walmart Apollo, LLC

Oracle International Corporation

Amit Jhunjhunwala

Lingfeng Zhang

Avinash Madhusudanrao Jade

Raghava Balusu

Srinivas Muktevi

Mangesh N Kulkarni Wadhonkar

Parul Aggarwal

Ashlin Ghosh

Rahul Kumar

Zhaoliang Duan

Srimant Misra

Satish Duggana

Benjamin R Ellison

Abhinav Pachauri

Mingquan Yuan

Han Zhang

William Craig Robinson

Raghav Mehta

Peeyush Taneja

Siddhartha Chakraborty

Feiyun Zhu

Tianyi Mao

William Craig Robinson, Jr

Anika Setia

Natasha Agarwal

Akshay Jindal

Ruofan Xu

Artur A Grochala

Nandini Tripathi

Zhiwei Huang

Rajiv Mishra

Systems and methods of creating reference template images for detecting and recognizing products at a product storage facility include an image capture device having a field of view that includes a product storage structure of the product storage facility, and a computing device including a control circuit and being communicatively coupled to the image capture device. The computing device obtains images of the product storage structure captured by the image capture device, analyzes the obtained images to detect individual ones of the products located on the product storage structure. Then, the computing device identifies the individual ones of the products detected in the images and crops each of the individual ones of the identified products from the images to generate cropped images. The computing device then creates a cluster of the cropped images, and selects one of the cropped images as a reference template image of an identified individual product.

Methods and systems for creating reference image templates for identification of products on product storage structures of a product storage facility

In some embodiments, apparatuses and methods are provided herein useful to processing captured images. In some embodiments, there is provided a system for processing captured images of objects at a product storage facility including a trained machine learning model; and a control circuit. The control circuit may group a plurality of product identifiers into one or more clusters based on at least one of visual similarity of corresponding images, textual similarity of corresponding associated descriptions, and associated relationships between product identifiers of the plurality of product identifiers; determine clusters having common elements that are at least within a similarity threshold of each other; merge the clusters with the common elements; and generate a mapping dataset used to retrain the trained machine learning model to identify a plurality of objects. The mapping dataset may include a plurality of associations of associated product identifiers to a single object.

Systems and methods for identifying different product identifiers that correspond to the same product

In some embodiments, apparatuses and methods are provided herein useful to processing captured images of objects at a product storage facility. In some embodiments, there is provided a system for processing captured images of objects including a trained machine learning model and a control circuit. In some embodiments, the trained machine learning model is configured to process unprocessed captured images. In some embodiments, the control circuit is configured to associate each of the processed images into one of a first group, a second group, or a third group; remove at least one processed image associated with the first group from the processed images in accordance with a first processing rule; and output remaining processed images associated with the first group and processed images associated with the second group to be used to retrain the trained machine learning model.

Systems and methods for processing images captured at a product storage facility

In some embodiments, apparatuses and methods are provided herein useful to updating inventory of products. In some embodiments, there is provided a system for updating inventory of products including a database; at least one image capture device; and a control circuit. The control circuit is configured to: process an image by: detecting units of one or more products in the image; grouping each detected unit into one or more clusters based on at least one of textual similarities, visual similarities, geometrical similarities, or relative spatial distance; detecting a product identifier for each cluster; identifying a product corresponding to the detected product identifier of each cluster; and counting the detected units associated with each cluster. The control circuit is configured to cause an update to the inventory of products in the database based on the counted detected units of each cluster.

Clustering of items with heterogeneous data points

In some embodiments, apparatuses and methods are provided herein useful to processing captured images. In some embodiments, there is provided a system for processing captured images of objects including a memory and a control circuit executing a trained machine learning model. The memory may be configured to store a plurality of images comprising first images and second images. The control circuit may be configured to: allocate each of the first images into one of a plurality of datasets; cluster each image in the dataset into one of a plurality of groups; select a sample from at least one of the plurality of groups; cluster each of the second images into one of dominant product identifier group and a non-dominant product identifier group; select a sample from the dominant product identifier group and a sample from the non-dominant product identifier group; and output the selected sample.

Systems and methods for analyzing and labeling images in a retail facility

Systems and methods of updating templates for use in recognizing individual products in images captured at a product storage facility include an image capture device that captures one or more images of product storage structure at a product storage facility, a computing device in communication with the image capture device, and an electronic database that stores keyword model templates and feature model templates associated with images of previously recognized individual products detected at the product storage facility. The computing device obtains the keyword and feature model templates associated with a recognized product from the electronic database, extracts the keywords from the products associated with the obtained keyword model templates, identifies products that are similar to the recognized product, and updates the keyword model template for each of the products to include must keywords and negative keywords, facilitating recognition of products in subsequent images captured by the image capture device.

Systems and methods of updating model templates associated with images of retail products at product storage facilities

In some embodiments, a system for remotely purchasing perishable products includes a plurality of image capture devices configured to capture images of perishable products, a product image database, a product attribute database, and a control circuit configured to execute computer program modules. An image processing module receives and processes images from the image capture devices to form a composite image. A product identification module identifies and segments the products in the composite image. A product selection module causes an application executed on a remote electronic device operated by a customer to display the composite image and receives from the application a selection of a product to purchase, at least one product attribute, and a quantity of individual product units to purchase. The product selection module identifies the individual product units having the selected product attribute(s) and outputs signaling to cause retrieval of the individual product units.

Systems and methods for remotely purchasing perishable products

Systems and methods of monitoring inventory of a product storage facility include an image capture device configured to move about the product storage areas of the product storage facility and capture images of the product storage areas from various angles. A computing device coupled to the image capture device obtains the images of the product storage areas captured by the image capture device and processes the obtained images of the product storage areas to detect individual products captured in the obtained images. Based on detection of the individual products captured in the images, the computing device analyzes each of the obtained images to extract meta data from the packaging the individual products to detect one more keywords and determine the locations of the detected keywords on the packaging, and then utilize this information to predict an identity of the products associated with the packaging.

Systems and methods of identifying individual retail products in a product storage area based on an image of the product storage area

In some embodiments, apparatuses and methods are provided to support individuals in a retail environment. Some embodiments, provide systems, comprising: a plurality of cameras positioned in a retail store; an image processing control circuit configured to determine a customer field of view; a product identification control circuit configured to identify a set of multiple products within the identified customer field of view, assign a viewing probability to each product without the first customer having to touch the first product; a product selection control circuit configured to determine a purchase probability of the first product; add the first product to a virtual electronic cart when the purchase probability of the first product exceeds a purchase probability threshold; and communicate instructions to control the worker personal computing device to instruct a worker to retrieve the first product.

Identifying object of interest for handicapped individuals based on eye movement patterns

In some embodiments, apparatuses and methods are provided herein useful to ensuring quality of meat cuts. In some embodiments, a system for ensuring quality of meat cuts comprises a capture device comprising an image capture device configured to capture an image of a cut of meat, a depth sensor configured to capture depth data, a transceiver configured to transmit the image and the depth data, a microcontroller configured to control the image capture device, the depth sensor, and the transceiver, a database configured to store meat cut specifications, and the control circuit configured to receive, from the capture device, the image and the depth data, retrieve, from the database, a meat cut specification, evaluate the image of the cut of meat and the depth data associated with the cut of meat, and classify the cut of meat.

Systems and methods for managing meat cut quality

In some embodiments, apparatuses and methods are provided herein useful to presenting packing instructions to a user. In some embodiments, a system for presenting packing instructions to a user comprises a database configured to store packing information and item visualizations, wherein the packing information includes placements of items in orders in containers, and wherein the item visualizations depict appearances of items sold by the retailer, a control circuit configured to receive an order identifier for an order, retrieve packing information associated with the order and item visualizations for the items in the order, generate a packing visualization for the order, wherein the packing visualization for the order depicts placement of the items in the order in the container, transmit, to a display device for presentation, the packing visualization for the order, and a display device configured to present the packing visualization for the order to the user.

Systems and methods for packing visualizations

In some embodiments, apparatuses and methods are provided herein useful to assess quality of produce at a facility. In some embodiments, there is provided a system for assessing quality of produce at a facility including a produce assessment station configured to provide a staging area to determine a quality classification of a target produce. By one approach, the produce assessment station includes a fixed surface; a rotatable base, a first arm comprising a microphone; a second arm comprising a tapping device; and a local control circuit. In one configuration, the local control circuit configured to rotate the rotatable base at a particular angle and at a particular time interval and receive audio data from the microphone. By one approach, the system includes a plurality of sensors and a portable device configured to provide a signal to the local control circuit to initiate the quality classification of the target produce.

System and method for assessing quality of produce

An aspect of the present disclosure facilitates third parties/server system to perform batch processing of requests requiring authorization from resource owners for repeat access to resources. In one embodiment, a server system/third party selects a next request from a batch of requests, with the next request requiring a protected resource (hosted on a second party) owned by a owner/user (first party). The server system checks whether an access token is present authorizing access of the protected resource by the server system on behalf of the owner. If the access token is not present, the server system communicates in an offline mode with the owner to receive the access token. The server system then processes the next request by accessing the protected resource using the present/received access token.

Facilitating third parties to perform batch processing of requests requiring authorization from resource owners for repeat access to resources

A system and method for error classification in an integration system. The system and method can receive a request to begin a transaction at the integration system. Additionally, the system and method can invoke an application associated with the request. In response to invoking the application, the system and method can receive a formatted error notification, the formatted error notification comprising at least one of a plurality of error formats. Additionally, the system and method can extract an error message text from the formatted error notification, resulting in a plain text of the formatted error notification. The system and method can also classify, at the error classifier, the formatted error notification based at least on training data and the plain text of the formatted error notification.

Growing community of inventors

Bengaluru, India

Amit Jhunjhunwala