Abstract: Methods and systems are disclosed for using generative machine learning models to generate fashion items for avatars. The methods and systems present a graphical user interface (GUI) comprising icons representing different types of avatar fashion items. The methods and systems receive input that selects an individual icon corresponding to an individual avatar fashion item. The methods and systems, in response to receiving the input, receive a selection of an individual texture from a texture selection region, the texture selection region comprising a first set of predefined textures and an option associated with a second set of textures generated based on a prompt by a generative machine learning model. The methods and systems apply the individual avatar fashion item with the individual texture to an avatar depicted in the GUI.

Abstract: Various implementations disclosed herein include devices, systems, and methods for providing situational awareness for HMD users. Some implementations involve providing a CGR experience to a first user wearing an HMD in a physical environment. Audio data or image data is obtained from a sensor at the physical environment during the CGR experience and an emotional state or direction of attention of a second person other than the first user is detected based on the audio data or image data. A notification or other change is provided in the CGR experience based on detecting the emotional state or direction of attention of the second person.

Abstract: An image displaying method including accepting from a user setting operation of setting a first surface of an object located in a real space as a plane onto which a projection image is virtually displayed, and displaying a first superimposed image in which a setting image showing a plane containing the first surface is superimposed on a captured image generated by capturing an image of the object.

Abstract: A vehicle-mounted monitoring device includes a main body detachably arranged on a vehicle, and the main body is arranged with a circuit board, a display screen, a first communication module and a second communication module, and the display screen, the first communication module and the second communication module are all electrically connected to the circuit board. The first communication module is configured to receive image data wirelessly transmitted from at least one camera and then transmit the image data to the circuit board, the second communication module is configured to receive tire pressure data wirelessly transmitted from at least one tire pressure sensor and then transmit the tire pressure data to the circuit board, and the circuit board transmits both the image data and tire pressure data to the display screen for display.

Abstract: A method includes obtaining a source image of a user and obtaining a driving video in which a face or a head of the user is moving. The method also includes generating metadata identifying animations to be applied to the source image so that the source image mimics at least some movements of the user's face or head in the driving video. The method further includes transmitting the source image and the metadata to an end user device configured to animate the source image based on the metadata. Generating the metadata includes suppressing one or more artifacts associated with one or more objects that temporarily occlude at least a portion of the user's head or body in the driving video or that temporarily appear in the driving video.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media that utilizes attention segregation loss and/or attention retention loss at inference time of a diffusion neural network to generate a text-conditioned image. In particular, in some embodiments, the disclosed systems utilize the attention segregation loss to reduce overlap between concepts by comparing attention maps for multiple concepts of a text query corresponding to a denoising step. Further, in some embodiments, the disclosed systems utilize the attention retention loss to improve information retention for concepts across denoising steps by comparing attention maps between different denoising steps. Accordingly, in some embodiments, by utilizing the attention segregation loss and the attention retention loss, the disclosed systems accurately maintain multiple concepts from a text query when generating a text-conditioned image.

Abstract: Systems, methods, and devices for data serialization extrusion of input images and raw input data. A method includes determining an input image comprising one or more lines and generating a bitmap based on the input image. The method includes performing edge detection on the bitmap to identify the one or more lines in the input image by applying a growth algorithm to pixels of the bitmap. The method includes calculating a pixel width of each of the one or more lines in the input image.

Abstract: An information processing device according to the present invention includes at least one memory and at least one processor which function as: an acquisition unit configured to acquire a first white-balance correction value corresponding to a white-balance correction value to be applied to a real-space image; a determination unit configured to determine, on a basis of the first white-balance correction value, a second white-balance correction value that is a white-balance correction value to be applied to a virtual-space image; a correction unit configured to correct the virtual-space image on a basis of the second white-balance correction value; and a superimposing unit configured to superimpose a virtual-space image after correction by the correction unit on the real-space image.

Abstract: A multimedia playback monitoring system includes an authentication image generator, a playback controller, an image compressor, and an analyzer. The authentication image generator is configured to generate a first transparent image according to a display terminal, and embed first identification information into the first transparent image to obtain an identity authentication image for a playback plan. The playback controller is configured to generate data to be displayed according to a received playback plan, a multimedia file corresponding to the playback plan, and the identity authentication image. The image compressor is configured to take and compress a screenshot of a target image displayed by the display terminal to obtain compressed data.

Abstract: The present disclosure relates to an avatar generator to generate an audio-visual avatar specific to an application, such as tutoring. The avatar generator includes a general synthesizer to receive a training dataset. The general synthesizer includes a voice synthesis module and a video synthesis module trained by the training dataset. The avatar generator includes a customized synthesizer consisting of a voice custom synthesis module and a video custom synthesis module trained on the audio-video samples of the target person. The avatar generator further includes a video generator to create an audio-visual avatar and is configured to synthesize a voice clone using an input text, process the voice clone, synthesize a video clone based on the video synthesis module and the video custom synthesis module, and apply the voice clone to the video clone.

Abstract: Systems, methods and computer-program products are provided for aiding in positioning between a surgical object and a machine vision camera in a surgical environment. The machine vision camera senses a position and orientation of a surgical object in its field of view. Controller(s) virtually define a zone within the field of view that indicates a range of acceptable positions for the surgical object relative to a position of the machine vision camera. The controller(s) obtain an acceptable orientation range for the surgical object relative to an orientation of the machine vision camera. The controller(s) execute a GUI to simultaneously present, on the display device, representations of: the field of view; the zone within the field of view; the position of the surgical object relative to the zone; and a current angle difference between the surgical object and the machine vision camera.

Abstract: A method includes executing a design generating program configured to display on a screen a graphical user interface having a virtual canvas window configured to present selected content objects to be incorporated into a unique design, and a content object selection window including available content objects to select from for presentation in the virtual canvas window. The program is configured to: receive, in the content object selection window, selection inputs indicating selection of first and second content objects and display the first and second content objects in the virtual canvas window. The program is further configured to receive, in the virtual canvas window, manipulation inputs to move the first content object to a target position relative to the second content object, and based on the first and second content objects overlapping, generate, for display in the virtual canvas window, the unique design.

Abstract: A method for training a controller to control a robotic system includes: receiving a neural network of an original controller for the robotic system based on origin data samples from an origin domain and labels in a label space, the neural network including encoder and classifier parameters, the neural network being trained to: map an input data sample from the origin domain to a feature vector in a feature space using the encoder parameters; and assign a label of the label space to the input data sample using the feature vector based on the classifier parameters; updating the encoder parameters to minimize a dissimilarity, in the feature space, between: origin feature vectors computed from the origin data samples; and target feature vectors computed from target data samples from a target domain; and updating the controller with the updated encoder parameters to control the robotic system in the target domain.

Abstract: System and methods facilitating optimized dynamic content loading are disclosed. To improve the user experience associated with loading dynamic content (e.g., content that adapts to user specific information), the system uses static content corresponding to the dynamic content to be presented to the user as the dynamic content loads. The system receives a user request to load a webpage associated with dynamic content and obtains static content associated with the dynamic content. The system then generates for display, the static content and then obtains the dynamic content corresponding to the static content. The system then generates for display, the dynamic content by displaying the dynamic content over the static content.

Abstract: The technology disclosed relates to a method of realistic rendering of a real object as a virtual object in a virtual space using an offset in the position of the hand in a three-dimensional (3D) sensory space. An offset between expected positions of the eye(s) of a wearer of a head mounted device and a sensor attached to the head mounted device for sensing a position of at least one hand in a three-dimensional (3D) sensory space is determined. A position of the hand in the three-dimensional (3D) sensory space can be sensed using a sensor. The sensed position of the hand can be transformed by the offset into a re-rendered position of the hand as would appear to the wearer of the head mounted device if the wearer were looking at the actual hand. The re-rendered hand can be depicted to the wearer of the head mounted device.

Abstract: This disclosure is directed to a method and apparatus for displaying an expression in a virtual scene. The method includes: displaying a virtual scene; displaying an expression selection region at a first target position in the virtual scene in response to a drag operation on an expression addition icon; and displaying the first target expression in the virtual scene in response to a selection operation on a first target expression in a plurality of first candidate expressions.

Abstract: A live camera feed may be analyzed to determine the identify of an object, and augmented reality overlay data may be determined based on that identity. The overlay data may include one or more tags that are each associated with a respective location on the object. The live camera feed may be presented on a display screen with the tags being positioned as the respective location.

Abstract: There are disclosed a method, apparatus, electronic device, and medium for rendering. The method for rendering comprises: determining colors of predetermined areas on a target object, the number of predetermined areas being at least a predetermined number, which is greater than 1; determining a target color superimposed to a target position according to the color of each predetermined area; superimposing the target color at the target position.

Abstract: A medical image processing apparatus includes the following. A processor is configured to: acquire a medical image obtained by capturing an image of an observation target; detect a region of interest from the medical image; and cause a display to display report information and the medical image, the report information reporting that the region of interest has been detected. A screen displayed on the display has a first region in which the medical image is displayed, and a second region provided outside the first region. The second region is divided into four regions by using line segments passing through a center of the first region, and a detected location of the region of interest is reported by displaying the report information in one of the four regions.

Abstract: A tag characterizing a portion of a multi-view interactive digital media representation (MVIDMR) may be determined by applying a grammar to natural language data. The MVIDMR may include images of an object and may be navigable in one or more dimensions. An object model location for the tag identifying a location within a three-dimensional object model may be determined by applying the grammar to the natural language data. The tag may then be applied to the MVIDMR by associating it with two or more of the images at positions determined based on the object model location.