Abstract: A computer-implemented method for generating custom content in a video game application is disclosed. The method includes receiving, via a first user interface, an input describing custom content such as a character or object. A model is used to determine a meaning of the input or a characteristic associated with a user profile. Based at least in part on the determined meaning or characteristic, the system generates the custom content for use within a virtual environment corresponding to execution of the video game application. The method further includes causing the generated custom content to be presented on a second user interface in connection with the execution of the video game application.

Abstract: An artificial intelligence (AI) model generates a control input for a virtual player of a video game based on input state data. A video stream of the video game is generated based on the control input for the virtual player of the video game. The video stream is transmitted to a client device. Training feedback data is received from the client device and includes a time-dependent series of human-generated feedback elements correlated to the video stream. Each feedback element indicates that the control input for the virtual player of the video game, as generated by the AI model, is either a player-approved control input or a player-disapproved control input. AI model training includes reinforcing use of the player-approved control input for the virtual player in response to the input state data, and diminishing use of the player-disapproved control input for the virtual player in response to the input state data.

Abstract: Methods and systems for warning misuse of a user account of an adult user includes tracking use of the user account. The interactions at the user account are monitored and when the content accessed by a user is adult content and the user is determined to be a child, providing an alert to the adult user informing the adult user of the child accessing age-inappropriate content.

Abstract: Systems and methods for generating chat-based user-generated content. The system may receive a chat-based communication over a communication network from a user device of a user and interpret the chat-based communication using a machine learning model. This may include identifying a meaning of content within the chat-based communication and predicting that the content within the chat-based communication indicates one or more virtual characteristics. A custom character for use within a virtual environment may be generated based on the identified meaning and the predicted virtual characteristics. The custom character may include the predicted virtual characteristics. The system may present the custom character to the user via a user interface of the user device.

Abstract: Methods and system for providing streaming content of a video game at a client device includes receiving frames of streaming content from a game server. The frames represent a current game state. The frames are analyzed to generate predicted frames that are likely to occur following the current frames. The predicted frames are stored in a prediction frame buffer and used to fill any gaps in subsequent frames representing subsequent game state received from the game server.

Abstract: To protect a user's privacy by reducing a malicious developer's ability to eavesdrop on unwitting HMD users by converting signals from a motion sensor in the HMD to speech or speaker recognition, a microphone can record ambient sound and voice which is subtracted from the motion sensor data before the sensor data is made available to the game developer. In another technique, a band pass filter subtracts frequency in the sensor signals within the voice range. Still a third technique blends statistical noise into the motion sensor signal before passing to game developers to obfuscate the user's speech. The amount by which voice components in the motion signal are eliminated or obfuscated can be tuned by a person or app.

Abstract: Techniques are described for smooth switchover of computer game control. The current states of game input is communicated to a new player assuming control. The new player is allowed time to catch up to the game. New player control is detected, and any errors are communicated to the new player. If there are differences between the old control scheme and that of the new player, they are reconciled. The outgoing player is adjusted to the transition.

Abstract: Important information with respect to image generation such as computer game video generation can be lost once the full image is composed. To avoid this, a customized capture script is created for a computer game to specify what to capture, for future use, in the middle of rendering a game video frame, such as frame buffer, depth buffer, stencil buffers, etc. The script also specifies when during frame rendering the data is to be recorded. By capturing data mid-frame, object boundaries on the screen and relative distances between objects can be inferred. This information can be used in 2D-to-3D conversion, object coloring for increased contrast, object highlighting, etc.

Abstract: To ensure that individuals gain as much as they can from a piece of text by adapting the text according to their specific reading needs and behaviors, metrics such as the position of a reader's gaze, pupil dilation, squinting behavior, fixation durations etc., are used to adapt the text according to reader's reading needs or habits and improve the reading experience. Machine learning may be used for this.

Abstract: Techniques for streaming (and non-streaming) XR applications that, when high latency is detected, automatically and naturally falls back on a cinematic experience in VR. For example, the game may be shown on a movie theater screen in a locally rendered environment where the game is played.

Abstract: Methods and system for providing streaming content of a video game at a client device includes receiving frames of streaming content from a game server. The frames represent a current game state. The frames are analyzed to generate predicted frames that are likely to occur following the current frames. The predicted frames are stored in a prediction frame buffer and used to fill any gaps in subsequent frames representing subsequent game state received from the game server.

Abstract: Methods and systems are provided for recording gameplay moments during gameplay of a video game is provided. The method includes executing the video game to generate a plurality of video frames. The method includes receiving inputs from a player of the video game. The inputs facilitate said gameplay of the video game by the player and causes updating of said plurality of video frames as the player makes progress in the video game. The method includes receiving an input to bookmark a frame region rendered in a video frame of said plurality of video frames. The method includes associating an identifier to the bookmark. The identifier is descriptive of content present in the frame region. The method includes saving state data for the bookmark. The state data for the bookmark enables subsequent selection of the bookmark from a user interface to load at least the video frame along with said identifier.

Abstract: Systems and methods of intelligent reporting within online communities are provided. A current communication session associated with a plurality of user device may be monitored. The current communication session may include a stream of audio-visual content generated in real-time based on interactions between the user devices. A recording trigger may be detected within the current communication session. A recording of a portion of the stream of audio-visual content may be recorded in response to the detected trigger event. The recording may be analyzed to attribute one or more sub-portions within the recording to one or more of the user devices. At least one of the sub-portions attributed to an identified one of the user devices may be determined to meet a moderation event. A report regarding the identified user device may be generated that includes the at least one sub-portion that meets the moderation event.

Abstract: A method performed for evaluating activity in a video game, including: executing a multi-player session of a video game; during the multi-player session, receiving flag event data from a first player device, the flag event data indicating that a first player has flagged a gameplay incident occurring during the multi-player session as potentially inappropriate; responsive to receiving the flag event data, then sending a request to a plurality of second player devices, wherein responsive to said request each of the plurality of second player devices renders a voting interface to obtain voting input from each of a plurality of second players regarding whether the gameplay incident is inappropriate; receiving said voting input from the plurality of second player devices, and responsive to said voting input identifying a threshold amount of the plurality of second players considering the gameplay incident to be inappropriate, then administering a penalty for the gameplay incident.

Abstract: Techniques that include presenting one or more images on a display device in response to execution of a computer program on a computing device. The techniques further include collecting gaze tracking information corresponding to a location of a user's gaze with respect to the one or more images presented on the display device. The techniques further include analyzing the gaze tracking information to detect a pattern of the user's gaze. The techniques further include upon detecting the pattern of the user's gaze, activating a first mode in which the user's gaze triggers commands affecting execution of the computer program.

Abstract: Systems and methods for customized dialogue support in virtual environments are provided. Dialogue maps stored in memory may specify dialogue triggers each associated with a corresponding dialogue instruction. Data regarding an interactive session associated with a user device may be monitored based on one or more of the stored dialogue maps. The presence of one of the dialogue triggers specified by the one or more dialogue maps may be detected based on the monitored data. Customized dialogue output may be generated in response to the detected dialogue trigger and based on the dialogue instruction corresponding to the detected dialogue trigger. The customized dialogue output may be provided to the interactive session in real-time with detection of the detected dialogue trigger.

Abstract: Techniques are described for smooth switchover of computer game control. The current states of game input is communicated to a new player assuming control. The new player is allowed time to catch up to the game. New player control is detected, and any errors are communicated to the new player. If there are differences between the old control scheme and that of the new player, they are reconciled. The outgoing player is adjusted to the transition.

Abstract: To protect a user's privacy by reducing a malicious developer's ability to eavesdrop on unwitting HMD users by converting signals from a motion sensor in the HMD to speech or speaker recognition, a microphone can record ambient sound and voice which is subtracted from the motion sensor data before the sensor data is made available to the game developer. In another technique, a band pass filter subtracts frequency in the sensor signals within the voice range. Still a third technique blends statistical noise into the motion sensor signal before passing to game developers to obfuscate the user's speech. The amount by which voice components in the motion signal are eliminated or obfuscated can be tuned by a person or app.

Abstract: A method, and system for controlling a computer program are disclosed. One or more images are presented on a display device in response to execution of the computer program on a computing device. Gaze tracking information corresponding to a location of a user's gaze with respect to the one or more images presented on the display device is collected over time. The gaze tracking information over time is analyzed to detect a pattern of the user's gaze characteristic of reading and upon detecting the pattern of the user's gaze characteristics of reading, a reading control mode is activated. In the reading control mode location and duration of the user's gaze triggers commands affecting execution of the computer program.

Abstract: A method, and system for controlling a computer program are disclosed. One or more images are presented on a display device in response to execution of the computer program on a computing device. Gaze tracking information corresponding to a location of a user's gaze with respect to the one or more images presented on the display device is collected over time. The gaze tracking information over time is analyzed to detect a pattern of the user's gaze characteristic of reading and upon detecting the pattern of the user's gaze characteristics of reading, a reading control mode is activated. In the reading control mode location and duration of the user's gaze triggers commands affecting execution of the computer program.