US 10,764,535 B1Granted

Facial tracking during video calls using remote control input

AssigneeFacebook, Inc.

InventorsBenjamin Powell, Yanni Wang, Yuan Chang, Tomas Brennessl

Life of this patent

Estimating the term…

Loading patent…

Abstract

A communication system enables users to select between individuals for tracking during video calls based on remote control input. The communication system establishes a video call session between a local client device and one or more remote client devices. The communication system uses a facial recognition algorithm to detect one or more faces from video data and obtains an identifier to each face. The communication system selects a first identifier. The communication system receives a navigation input from a remote control coupled to the communication system and, based on the input, selects a second identifier. The communication system receives an operation input and applies a center of focus of the video data to a second face corresponding to the second identifier.

Claims12

Dependent claims

A method comprising:

establishing a video call session between a local client device and one or more remote client devices;
and
during the video call session:
detecting, by a facial detection algorithm, one or more faces from video data associated with the video call session;
assigning an ordered identifier for each face of the one or more faces, the ordered identifiers ordered according to respective times of detection of the one or more faces;
transmitting video data associated with the video call session for display on a display device;
selecting a first ordered identifier associated with a first face;
receiving, from a remote control device communicatively coupled to the local client device, a navigation input for cycling selection of the one or more faces, the cycling based on the ordered identifiers;
and
responsive to the navigation input, selecting a second ordered identifier, associated with a second face;
receiving an operation input to perform an operation when the second ordered identifier is selected;
and
performing the operation on the video data with respect to the second face.

10 more elements

A non-transitory computer-readable storage medium storing computer program instructions executable by a processor to perform operations comprising:

establishing a video call session between a local client device and one or more remote client devices;
and
during the video call session:
detecting, by a facial detection algorithm, one or more faces from video data associated with the video call session;
assigning an ordered identifier for each face of the one or more faces, the ordered identifiers ordered according to respective times of detection of the one or more faces;
transmitting video data associated with the video call session for display on a display device;
selecting a first ordered identifier associated with a first face;
receiving, from a remote control device communicatively coupled to the local client device, a navigation input for cycling selection of the one or more faces, the cycling based on the ordered identifiers;
and
responsive to the navigation input, selecting a second ordered identifier, associated with a second face;
receiving an operation input to perform an operation when the second ordered identifier is selected;
and
performing the operation on the video data with respect to the second face.

10 more elements

Dependencies are inferred from the claim text.