Abstract: A method and an apparatus for processing a 3D scene are presented. Techniques are disclosed for determining light source locations, including tracking a current viewpoint of a camera capturing object(s) in a 3D scene and determining a reference viewpoint relative to the current viewpoint of the camera. According to aspects, a light source location is determined by obtaining a registered map of real cast shadows of the object(s) from an input image captured by the camera, registered with respect to the reference viewpoint. Then, for candidates of light sources, obtaining respective maps of virtual shadows of the object(s) created with respect to the reference viewpoint, and determining the location of the light source based on the candidates of light sources with respective maps of virtual shadows that match the registered map of real cast shadows.

Abstract: One variation of a method for detecting and visualizing objects within a space includes: retrieving a map of the space annotated with known locations of target anchor objects and regions in the space; accessing a set of images annotated with object types and locations of objects captured by a set of sensor blocks; projecting the set of images onto the map to form a visualization representing objects in the space based on known locations of the set of sensor blocks; isolating a target anchor object at a known location in a region of the visualization; detecting a mutable object at a location in the region; calculating an offset distance between the location and the known location; and, in response to the offset distance exceeding an offset distance threshold, highlighting the mutable object in the visualization as a deviation; and generating a notification to investigate the mutable object in the region.

Abstract: To present augmented reality features without localizing a user, a client device receives a request for presenting augmented reality features in a camera view of a computing device of the user. Prior to localizing the user, the client device obtains sensor data indicative of a pose of the user, and determines the pose of the user based on the sensor data with a confidence level that exceeds a confidence threshold which indicates a low accuracy state. Then the client device presents one or more augmented reality features in the camera view in accordance with the determined pose of the user while in the low accuracy state.

Abstract: Embodiments of this application disclose an image style conversion method performed by an electronic device. The method includes: performing quality enhancement on a first target style image to obtain a second target style image; performing feature extraction on the second target style image to obtain a target style feature; performing migration training on a preset target style conversion model by using a full style conversion model and the target style feature to obtain a target style conversion model; inputting a full style feature, the target style feature, and a to-be-converted image into the target style conversion model, and performing style conversion on the to-be-converted image using the target style conversion model to obtain a target image conforming to a target style.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for processing multimodal content to generate summaries or responses using a multimodal large language model. In one or more embodiments, the disclosed systems the disclosed systems utilize the multimodal large language model to generate various types of synthesized responses corresponding to multimodal content items that contain data and information within images. For example, in some embodiments, in response to receiving a request to generate a synthesized response corresponding to a multimodal content item, the disclosed systems employ preprocessing pipelines that generate thumbnail images from the multimodal content item and use the thumbnail images to generate a data structure for a prompt for the multimodal large language model.

Abstract: An information processing system comprises processing circuitry configured to display a first user interface element on a screen of a virtual space, the first user interface element for starting an application browser; start the application browser in a case that an operation is performed on the first user interface element by a user; display a second user interface element which shares information displayed by the application browser; send a sharing request to a server in a case that an operation is performed on the second user interface element by the user, the sharing request being a request to share information displayed by the application browser with a device used by a different user viewing the screen of the virtual space; and display, as shared information, information displayed by the application browser on a display object disposed in the virtual space.

Abstract: An image processing method includes an electronic device configured to perform style migration processing on a first image sequence based on a target migration style by using a fused style migration model into which a plurality of single-style migration models is fused in order to obtain a second image sequence. A style of a 1st frame of image to a style of a last frame of image in the second image sequence change in a first style order in styles of output images of the plurality of single-style migration models. The first image sequence may be from a video shot by using the electronic device. The electronic device may save a plurality of frames of images in the second image sequence as a video. The video may present an effect of rapid time lapse during play.

Abstract: Systems and methods for updating remote memory side caches in a multi-GPU configuration are disclosed herein. In one embodiment, a graphics processor for a multi-tile architecture includes a first graphics processing unit (GPU) having a first memory, a first memory side cache memory, a first communication fabric, and a first memory management unit (MMU). The graphics processor includes a second graphics processing unit (GPU) having a second memory, a second memory side cache memory, a second memory management unit (MMU), and a second communication fabric that is communicatively coupled to the first communication fabric. The first MMU is configured to control memory requests for the first memory, to update content in the first memory, to update content in the first memory side cache memory, and to determine whether to update the content in the second memory side cache memory.

Abstract: A method includes receiving, via one or more processors, video data and audio data associated with respective participants in a video session, determining, via the one or more processors, words spoken by a speaker participant in the video session based on the audio data, determining, via the one or more processors, a location of the speaker participant in a framing of the video data based on the video data and the audio data, generating, via the one or more processors, an animated avatar to provide sign language representing the words spoken by the speaker participant, modifying, via the one or more processors, the video data of the video session to include the animated avatar based on the location of the speaker participant, and outputting, via the one or more processors, the video data that includes the animated avatar.

Abstract: In an approach to improve adaptive content and responsive web content, embodiments of create a profile associated with a user based on monitoring online activity of the user and analyze and classify content displayed to the user based on collected data associated with a reaction from the user. Further, embodiments derive a preliminary set of topics of interests based on the reaction of the user and the analyzed and classified content and associate the preliminary set of topics of interest with the reaction of the user and the analyzed and classified content. Additionally, embodiments, analyze a location of interest to the user on associated with the content based on the preliminary set of topics of interests to personalize a display of the location for interaction by the user, and present, by a user interface, a personalized display of the location to the user.

Abstract: A method and system for directing image rendering, implemented in a computer system including a plurality of processors includes determining one or more processors in the system on which to execute one or more commands. A graphics processing unit (GPU) control application program interface (API) determines one or more processors in the system on which to execute one or more commands. A signal is transmitted to each of the one or more processors indicating which of the one or more commands are to be executed by that processor. The one or more processors execute their respective command. A request is transmitted to each of the one or more processors to transfer information to one another once processing is complete, and an image is rendered based upon the processed information by at least one processor and the received transferred information from at least another processor.

Abstract: An information processing device includes: a first acquisition unit configured to acquire a capturing time of a lesion image instructed to be saved by a user, from a series of images captured by an endoscope during examination with the endoscope; a second acquisition unit configured to acquire a capturing time of a lesion image detected by detection processing for the series of images captured by the endoscope during the examination; and a display control unit configured to cause a display device to display a first capturing time and a second capturing time which are plotted on a time axis, the first capturing time being the capturing time acquired by the first acquisition unit, the second capturing time being the capturing time acquired by the second acquisition unit.

Abstract: A method, including converting an audio input to a text input, generating, by an artificial intelligence (AI) persona, at least one of a visual output and an audio output based on the text input and a face input, synchronizing, by the AI persona, at least one of the visual output and the audio output to movement of a rendered facial image, and modifying, by the AI persona, at least one of the visual output and the audio output in real-time in response to real-time changes in the audio input and the face input.

Abstract: A computer-implemented method that includes (1) maintaining access to a database that contains object augments mapped to one or more objects along with which the object augments are configured to be presented to and sensed by a user via an extended-reality system, (2) detecting an object in the user's environment that is mapped to at least one object augment in the database, (3) determining whether a presentation condition associated with the at least one object augment is satisfied, (4) prioritizing presentation of the at least one object augment to the user via the extended-reality system when the presentation condition is satisfied, and (5) constraining presentation of the at least one object augment to the user via the extended-reality system when the presentation condition is not satisfied. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: An information processing device includes: a first acquisition unit configured to acquire a capturing time of a lesion image instructed to be saved by a user, from a series of images captured by an endoscope during examination with the endoscope; a second acquisition unit configured to acquire a capturing time of a lesion image detected by detection processing for the series of images captured by the endoscope during the examination; and a display control unit configured to cause a display device to display a first capturing time and a second capturing time which are plotted on a time axis, the first capturing time being the capturing time acquired by the first acquisition unit, the second capturing time being the capturing time acquired by the second acquisition unit.

Abstract: A camera tracking system is disclosed for computer assisted navigation during surgery. The camera tracking system is configured to identify a reference array tracked by a set of tracking cameras attached to an extended reality (XR) headset, and determine whether the reference array is registered as being paired with characteristics of one of a plurality of surgical tools defined in a surgical tool database. The camera tracking system is further configured to, based on the reference array being determined to not be registered and receiving user input, register the reference array to be paired with characteristics of one of the plurality of surgical tools selected based on the user input. The camera tracking system is further configured to provide a representation of the characteristics to a display device of the XR headset for display to the user.

Abstract: Methods and systems disclosed herein describe creating and/or distributing media files associated with a playbook for a sports team. A user may use a computing device to create, or edit, a play. The location of the players, the ball, and their respective movements may be recorded in a flat file that is used to generate an animation of the play. The flat file may be transmitted to a server, which may use the flat file to generate the animation of the play. The animation of the play may be distributed (e.g., sent) to one or more people associated with the team. Additionally, the animation of the play may be added to a playbook of a sports application, which may be a single point of contact for coaches, managers, parents, and/or athletes.

Abstract: A computer-implemented method of generating multimodal data. The method comprises using a token generation neural network to generate, autoregressively, an output sequence of multimodal tokens, and in response to a next multimodal token being a start-of-image token, generating an image using an image generation subsystem conditioned on features representing the current sequence of multimodal tokens obtained from the token generation neural network. The method further comprises processing the image to convert pixels of the image into a sequence of image tokens, each image token comprising a block encoding of values of the pixels in a different region of the image that maps a set of values of the pixels to a respective image token, and appending the sequence of image tokens to the current output sequence of multimodal tokens as the next multimodal tokens in the output sequence of multimodal tokens.

Abstract: Described is a system for improving machine learning models. In some cases, the system improves such models by identifying a performance characteristic for machine learning model blocks in an iterative denoising process of a machine learning model, connecting a prior machine learning model block with a subsequent machine learning model block of the machine learning model blocks within the machine learning model based on the identified performance characteristic, identifying a prompt of a user, the prompt indicative of an intent of the user for generative images, and analyzing data corresponding to the prompt using the machine learning model to generate one or more images, the machine learning model trained to generate images based on data corresponding to prompts.

Abstract: The embodiment of the invention provides a method and apparatus for image generation, an electronic device and a storage medium. The method includes: in response to a triggering operation for image effect processing, performing effect processing on a to-be-processed image to obtain an effect image; in response to a style type of the effect image being the same as a used style type of an effect image generated last time, determining at least one to-be-selected style type based on a predetermined style probability distribution; determining a target style type based on the at least one to-be-selected style type; and in response to the triggering operation for the image effect processing being detected, processing the to-be-processed image into an effect image matching the target style type.