Abstract: Methods, systems, and code that can provide enhancements to content recorded during a game session of a game program. An example can provide a narration or other enhancement to the content. The narration can be added to a video segment of a game session of a game program.

Abstract: Systems and methods are disclosed for expanding the output modalities of a gameplay session. Gameplay input data, such as audio data, video data, haptics data, and Gameplay context data, can be processed using machine learning (ML) models to generate output state vectors, which can be used to drive state changes in remote devices, such as smarthome devices (e.g., smart lightbulbs, smart light switches, smart fans, smart thermostats, smart blinds, smart speakers, etc.). To generate the output state vectors, first gameplay features are extracted from the gameplay input data, then the gameplay features are processed by a state vector generation model to generate the state vectors. Each modality of gameplay input data can be processed by a corresponding feature extractor. Gameplay sessions can be evaluated to train the ML models.

Abstract: Described herein are techniques for implementing a virtualized physical controller. The techniques may comprise receiving. from a user via a touchscreen display. a first touch input corresponding to at least one input mechanism of a set of input mechanisms. determining a first state that is currently active. generating. based on the first state. first activation data that includes an indication of a first action. The techniques may further comprise receiving. from the user via the touchscreen display. a second touch input corresponding to the at least one input mechanism. determining a second state that is currently active, and generating. based on the second state, second activation data that includes an indication of a second action different from the first action.

Abstract: Methods for tagging content in a video game title during game play executed on a game cloud system, and social sharing methods. One method includes receiving a pause indication during the game play and suspending the game play to hold the video game title in a current frame image, storing the game state, and receiving the tag data that is associated to a location in the current frame image and to a user identifier for the current frame image. The method further includes generating a video recording of the game play having the tag data and generating a mini-game from a portion of the video game title. The mini-game includes a playable portion to enable view of a portion of the video recording for the mini-game and an executable portion to enable the play of game code for the mini-game that is a portion of the video game title.

Abstract: There has been provided a method of modifying a content stream provided to a user electronic device, the method comprising receiving device data from the user electronic device, processing the device data to determine whether one or more network conditions have been met, and modifying the content stream and/or one or more settings associated with either the content stream and or a game server. There has also provided a method of rendering a content stream provided by a sender, the method comprising, sending device data, receiving a modified content stream, and rendering the modified content stream.

Abstract: System, process, and device configurations are provided for identifying depth of gaming elements. A method can include detecting at least one game object in video data, wherein the at least one game object is a display element and determining a depth parameter for the at least one game object, wherein the depth parameter is a distance value of the at least one game object relative to at least one background display element for the video data. The video data may be output using the depth parameter for the at least one game object. Embodiments can create title specific data from analysis of rendered video data to determine modifications for adding depth and modifying rendering of titles. Depth determinations can include features to allow for different rendering styles and presentation in a three-dimensional format or by a headset. Processes and configurations are also provided for generating contextual information about game objects.

Abstract: In some implementations, a method includes receiving, at a server, environment data. The method further includes providing the environment data to one or more machine-learning models. The method further includes receiving one or more candidate camera settings from the one or more machine-learning models, where the one or more candidate camera settings are based on the environment data. The method further includes displaying the one or more candidate camera settings in a user interface of a camera device, where the one or more candidate camera settings guide a user in improving a quality level of one or more images captured by the camera device. The method further includes modifying the one or more camera settings of the camera device based on the one or more candidate camera settings.

Abstract: Machine learning is used to generate a first mipmap of a texture having a first compression based on a second mipmap of the same texture and having a second compression without using compression or decompression in generating the first mipmap. The first mipmap can then be used to render a computer graphics object. Training can be done using as input decompressed blocks from originally compressed blocks.

Abstract: Methods and systems are provided for providing a player with assistance during gameplay of a video game is provided. The method includes executing a video game for a game session. The player controls a player character during the game session. The method includes detecting selection of ghost assistance for the game session. The method includes linking or connecting the game session to an assistance artificial intelligence (AI) engine. The assistance AI engine operates to provide ghost assistance to the player in accordance with a mode setting. The ghost assistance is provided via a ghost character. The method includes rendering, during the game session, the ghost character performing interactive actions in the game. The interactive actions are generated responsive to control inputs generated by the assistance AI engine. The interactive actions by the ghost character are configured to progress the ghost character along an interactive path of the game.

Abstract: System, process and device configurations are provided for analyzing and modifying a game environment. A method can include generating at least one game scenario for an electronic game including control settings for an audio configuration, a video configuration and a gameplay configuration. The method may include outputting the at least one game scenario using the control settings, and detecting user interest for the at least one game scenario. User interest for the at least one game scenario may be detected using a kernel function including a kernel element for each of a visual style, an audio style and a gameplay style. User profile data may be updated for the at least one user based on detected user interest for the at least one game scenario. Output of a game state may be controlled for the electronic game using the user profile.

Abstract: Systems, processes and device configurations are provided for analyzing and validating image data. Processes include detecting image data, controlling analysis to perform object recognition of a first object in the image data and an operation to determine a characteristic of the first object using the image data and controlling at least one validation operation to verify the characteristic of the first object, the validation operation generating at least one validation output. Processes can also include controlling output of the at least one validation output, wherein the output of the at least one validation operation includes output of at least one graphical element with the image data. Processes and configurations are also provided for labeling and identifying one or more edited image data, generated image data and image data relying too much on seed material.

Abstract: Described herein are techniques for implementing dynamic movement control on a virtualized physical controller, comprising receiving, on a touch-screen display, an indication of a first touch input received from a user at a first location, determining, based at least in part on a distance between the first location and a second location associated with an origin point for an input mechanism being less than a threshold distance value, that a persistent directional input has been initiated, maintaining the persistent directional input as long as touch input is continuously detected, and while the persistent directional input is maintained: monitoring a touch input vector associated with the persistent directional input, upon detecting an angular change in the touch input vector that is greater than a threshold change value, determining that a reversal input has been provided, and causing the avatar to perform a reversal action corresponding to the reversal input.

Abstract: Non-players can vote and comment on uploaded video clips or livestreams of video games played by video game players. The video game players that upload the clips or do the livestreaming and that then receive the most votes can be awarded digital awards. The non-players providing the best comments can also be awarded digital awards. This incentivizes these behaviors, providing ample training data in the process so that the votes and comments can be used to train a model to make inferences related to video game video. For instance, the model may be trained to provide auto-generated comments in real time as a second video game is played, where the comments are in video game domain-specific language.

Abstract: Described herein are techniques for transmitting state data between a controller and another electronic device. In embodiments, such techniques may comprise receiving, at a user device, input that relates to a first action to be performed by an avatar; providing, to a second electronic device, a first set of state data generated from the received input over a first communication channel; and determining a user intent that relates to a second action to be performed by the avatar. Based on the user intent, a second set of state data can be generated that is different from the first set of state data and includes less than all available state data. The second set of state data can be provided to the second device over a second communication channel, causing the avatar to perform a modified action.

Abstract: Generating custom team emblems using stable diffusion based on input text describing a desired image. A circle is overlaid in the center of a pure-color background representing each team's “color” and used as the input to stable diffusion img2img to produce emblems. This produces high-quality emblem outputs that generally match the input color.

Abstract: A micro-OLED display includes a substrate with a coating made of IR-transparent material such as Zinc Selenide or Zinc Sulfide to allow a pupil tracking camera, which may include photodiodes, to be mounted directly behind the display. The coating is transparent to near infrared (NIR) radiation but opaque to visible light.

Abstract: In one aspect, schemas for interactive content creation may be created using large language models (LLMs). The schemas may refer to one another and may be used to build schema instances off the schemas themselves. The schema instances may also refer to each other. Computer code for actions in the interactive content can then be written in a given programming language to refer to the schemas and schema instances, enabling quick code creation with less chances of crashing due to the strongly-typed language being used, including the schema references to each other. The interactive content might be video game content or another type of interactive content.

Abstract: A text-to-3D machine learning (ML) model is trained by rendering a pre-existing 3D model along with an associated depth map using plural camera settings and rendering a neural radiance field (NeRF) along with an associated depth map using the plural camera settings used in rendering the 3D model. A composite is established from the depth maps, which is used to train the ML model.

Abstract: A method including collecting one or more inputs, each of which describes a target object. The method including generating a plurality of images of the target object using an image generation artificial intelligence system configured for implementing latent diffusion based on the one or more inputs. The method including decomposing the target object into a first plurality of attributes based on the plurality of images of the target object, wherein each of the plurality of attributes includes one or more variations. The method including receiving selection of one or more of a plurality of variations of the plurality of attributes. The method including blending the one or more of the plurality of variations of the plurality of attributes that have been selected into one or more options of the target object.

Abstract: Systems and methods for providing extended reality (XR) interactions are provided. A system includes a motion tracker device, an XR display device, and a processor. The processor is configured to execute computer-readable instructions to display XR content in a three-dimensional (3D) space for an interaction session, track the full-body motion of a user relative to the 3D space, and define an interaction plane in the 3D space. The XR content includes obstacle elements and target elements moving toward the interaction plane in the 3D space. The system determines an event score for an interaction event based on checks whether a user body part overlaps the target element in the interaction plane and whether all user body parts avoided overlapping the obstacle element in the interaction plane.