Generating an image mask using machine learning

US10402689B1 · kind B1 · utility

Assignee

• SNAP INC. · US

Inventors

• Lidiia Bogdanovych · Los Angeles, US
• William Brendel · Los Angeles, US
• Samuel Edward Hare · Los Angeles, US
• Fedir Poliakov · Odesa, UA
• Guohui Wang · Los Angeles, US
• Xuehan Xiong · Pittsburgh, US
• Jianchao Yang · Santa Monica, US
• Linjie Yang · Los Angeles, US

Key dates

• Filing date: Sep 15, 2017
• Grant date: Sep 3, 2019
• Expiry date: Feb 16, 2038

Classification

• Technology area (CPC H): Electricity
• CPC primary: H04N7/141
• WIPO field: Audio-visual technology
• WIPO sector: Electrical engineering

Abstract

A machine learning system can generate an image mask (e.g., a pixel mask) comprising pixel assignments for pixels. The pixels can be assigned to classes, including, for example, face, clothes, body skin, or hair. The machine learning system can be implemented using a convolutional neural network that is configured to execute efficiently on computing devices having limited resources, such as mobile phones. The pixel mask can be used to more accurately display video effects interacting with a user or subject depicted in the image.