Amazon Technologies, Inc.

Microsoft Technology Licensing, LLC

Pranab Mohanty

Ammar Chinoy

Joachim Sebastian Stahl

Gianna Lise Puerini

Jason Michael Famularo

Frank Florian Liberato, Jr

Jared Joseph Frank

Shrivatsan Vasudhevan

Emilio Ian Maldonado

David Allen Smith

Jiajun Zhao

Maxwell Aaron Corman

Yasser Baseer Asmi

Sudarshan Narasimha Raghavan

Prithviraj Banerjee

Intaik Park

Kieran Brantner-Magee

Lucas Lin

Saikat Sen

Korhan Ileri

Described is a multiple-camera system and process for re-identifying a user located in a materials handling facility based on user patterns and/or descriptors representative of the user. In one implementation, a user pattern and/or a plurality of descriptors representative of a user are maintained as a position of a user is tracked through a materials handling facility. If the tracking of the user is lost, the last known position is stored with the user pattern and/or descriptors. If a new object is detected and confirmed to be a user, a user pattern and/or descriptors of the new object are compared with the stored user pattern and/or descriptors to determine if the new object is the user.

Updating a position of a user within a materials handling facility

Markers may be presented on different objects in a facility, such as shelves or totes. A marker can use color, placement of elements, blending effects, or infrared light layers to encode data about the object. During operation, a camera within the facility may acquire an image, which includes an image of the marker as presented on the object. The marker can be analyzed to detect marker data encoded therein. By decoding the marker data, details about the marker can be extracted, such as an identification of the object. In addition, using details about the marker location within the image and location data associated with the camera, a location of the object relative to the camera can be determined, as well as an orientation of the object relative to the camera. As the object moves within the facility, the object can be tracked based on this location and orientation data.

Tracking of objects using markers

Described is a multiple-camera system and process for detecting a user within a materials handling facility and tracking a position of the user as the user moves through the materials handling facility. In one implementation, a plurality of depth sensing cameras are positioned above a surface of the materials handling facility and oriented to obtain a top-down view of the surface of the materials handling facility, along with any objects (e.g., users) on the surface of the materials handling facility. The depth information from the cameras may be utilized to detect objects on the surface of the materials handling facility, track a movement of those objects and determine if those objects are users.

Collecting user pattern descriptors for use in tracking a movement of a user within a materials handling facility

Machine-readable optical markers may be presented on a surface of a shelf, on a display device, or on a structure. Markers are presented as an array of graphical elements arranged in rows and columns. The graphical elements in the array of the mark utilize color and placement to encode marker data. In addition, a source image graphic may also be blended with a portion of the graphical elements in the array to create a blended image. In other embodiments, the marker may be presented with one or more borders surrounding the perimeter of the graphical elements in the array. The marker data may be presented using infrared light with a visible light layer presented over the array of graphical elements.

Presentation of machine-readable optical markers

Re-establishing tracking of a user within a materials handling facility

Segmenting a user pattern into descriptor regions for tracking and re-establishing tracking of a user within a materials handling facility

Described are systems and techniques for using machine-readable optical markers to calibrate a machine vision system. In one implementation, the machine vision system may generate calibration data associating spatial coordinates with image coordinates of the machine-readable marker in an image acquired by a camera. The machine-readable marker may comprise a blend of an optical code that encodes marker data and a human-readable source graphic image. In some implementations, the optical code of the markers may be detectable under infrared light while remaining invisible to the unaided eye.

Machine vision calibration system with marker

Examples for detecting and removing non-natural language within natural language to enhance performing content analysis on the natural language are provided herein. A plurality of terms is identified in a phrase, and a sliding window having a defined length is placed over a first sequence of terms from the plurality of terms. The first sequence of terms includes a first term, a second term, and a third term, the first term and the third term being adjacent to the second term. Based on the first term, the second term, and the third term, a determination is made as to whether the second term represents non-natural language. Upon determining that the second term is non-natural language, the second term is labeled as non-natural language and is removed from the plurality of terms based on determining the second term as non-natural language.

Identifying non-natural language for content analysis

Growing community of inventors

Redmond, WA, United States of America

Pranab Mohanty