Techniques for training and using machine learning models for video encoding mode selection are described. According to some embodiments, a computer-implemented method includes receiving a live video at a content delivery service, extracting one or more features for a plurality of macroblocks of a frame of the live video, determining an encoding mode from a plurality of encoding modes for each of the plurality of macroblocks of the frame with a machine learning model based at least in part on an input of the one or more features, performing a real time encode of the frame of the live video based at least in part on the determined encoding modes to generate an encoded frame by the content delivery service, and transmitting the encoded frame from the content delivery service to a viewer device.

Techniques for training and using machine learning models for video encoding mode selection are described. According to some embodiments, a computer-implemented method includes receiving a live video at a content delivery service, extracting one or more features for a plurality of macroblocks of a frame of the live video, determining an encoding mode from a plurality of encoding modes for each of the plurality of macroblocks of the frame with a machine learning model based at least in part on an input of the one or more features, performing a real time encode of the frame of the live video based at least in part on the determined encoding modes to generate an encoded frame by the content delivery service, and transmitting the encoded frame from the content delivery service to a viewer device.

Systems and methods are directed to automatically creating customized whiteboard backgrounds. A network system accesses metadata associated with a virtual presentation (e.g., title, topic, tenant identifier). First image data is identified based on first data of the metadata and second image data is identified based on second data of the metadata. Using the first image data and the second image data, the network system generates a plurality of whiteboard backgrounds by combining a first object obtained from the first image data with a second object obtained from the second image data to form each whiteboard background. The network system then causes presentation of a representation of each of the plurality of whiteboard backgrounds on a user interface of a host, who can select one of the representations. In response to receiving a selection, a whiteboard background corresponding to the selected representation is displayed as background on a whiteboard canvas.

The present invention relates to a method for dynamic image content replacement in a video stream comprising generating a set of key image data (K) comprising a sequence of at least two different key images (K1, K2), periodically displaying said set of key image data (K) on a physical display, generating at least a first original video stream (O1) of a scene which includes said physical display by recording said scene with a camera, wherein said at least one video stream (O1) comprises key video frames (FK1, FK2), captures synchronously with displaying each of said at least two different key images (K1, K2) of said set of key image data (K) on said physical display, generating a mask area (MA) corresponding to an active area of said physical display visible in said key video frames from differential images (AFK) obtained from consecutive key video frames (FK1, FK2), generating at least one alternative video stream (V) by inserting of alternative image content (I) into the mask area (MA) of an original video stream, and broadcasting at least said at least one alternative video stream.

The present invention relates to a method for dynamic image content replacement in a video stream comprising generating a set of key image data (K) comprising a sequence of at least two different key images (K1, K2), periodically displaying said set of key image data (K) on a physical display, generating at least a first original video stream (O1) of a scene which includes said physical display by recording said scene with a camera, wherein said at least one video stream (O1) comprises key video frames (FK1, FK2), captures synchronously with displaying each of said at least two different key images (K1, K2) of said set of key image data (K) on said physical display, generating a mask area (MA) corresponding to an active area of said physical display visible in said key video frames from differential images (AFK) obtained from consecutive key video frames (FK1, FK2), generating at least one alternative video stream (V) by inserting of alternative image content (I) into the mask area (MA) of an original video stream, and broadcasting at least said at least one alternative video stream.

Provided are a method for pushing anchor information, a computer device, and a storage medium. The method for pushing the anchor information includes: recalling an anchor user; determining an anchor type of the anchor user by using a live streaming history and a live streaming efficiency as classification dimensions; calculating an interaction score of the anchor user, wherein the interaction score represents a feature of a viewer user viewing a live streaming room of the anchor user; calculating a comprehensive score of the anchor user based on the interaction score and the anchor type; and pushing anchor information of the anchor user to the viewer user based on the comprehensive score.

Provided are a method for pushing anchor information, a computer device, and a storage medium. The method for pushing the anchor information includes: recalling an anchor user; determining an anchor type of the anchor user by using a live streaming history and a live streaming efficiency as classification dimensions; calculating an interaction score of the anchor user, wherein the interaction score represents a feature of a viewer user viewing a live streaming room of the anchor user; calculating a comprehensive score of the anchor user based on the interaction score and the anchor type; and pushing anchor information of the anchor user to the viewer user based on the comprehensive score.

A student terminal is for viewing a class given in a virtual space that is immersive. The student terminal includes: a VR function section configured to display the virtual space according to virtual space information; and an input section for receiving a video capturing a desk of a student who views the class. The VR function section extracts, from the video, an area including a top plate of the desk corresponding to a desk object in the virtual space, and performs image composition for fitting a video capturing the area onto a top plate of the desk object.

A method includes: receiving text data and motion data of a virtual object, the motion data including a motion identifier of a specified motion and a start position identifier of a start position that the specified motion starts being in line with text in the text data; generating audio data and expression data of the virtual object according to the text data, and generating facial images of the virtual object according to the expression data; generating a background image sequence containing the specified motion according to the start position identifier and the motion identifier, the background image sequence including at least one background image; performing image fusion processing on the facial images and the at least one background image to obtain one or more live video frames; and synthesizing the live video frames with the audio data into a live video stream in real time.

A method includes: receiving text data and motion data of a virtual object, the motion data including a motion identifier of a specified motion and a start position identifier of a start position that the specified motion starts being in line with text in the text data; generating audio data and expression data of the virtual object according to the text data, and generating facial images of the virtual object according to the expression data; generating a background image sequence containing the specified motion according to the start position identifier and the motion identifier, the background image sequence including at least one background image; performing image fusion processing on the facial images and the at least one background image to obtain one or more live video frames; and synthesizing the live video frames with the audio data into a live video stream in real time.