Disclosed are methods, systems and apparatuses for creating an animal-shaped avatar using a human face. An avatar creation method according to example embodiments includes analyzing an image including a human face and automatically creating an animal-shaped avatar corresponding to the human face.

A method for capturing real-world information into a virtual environment is provided. The method is suitable for a head-mounted device (HMD) located in a physical environment, and includes the following operations: providing the virtual environment, wherein a real-world content within the virtual environment is captured from a part of the physical environment corresponding to a perspective of the HMD; tracking a feature point which located within the physical environment and moved by a user, so as to define a selected plane of the real-world content by projecting a moving track of the feature point onto the real-world content; capturing image information corresponding to the selected plane; and generating a virtual object having an appearance rendered according to the image information, in which the virtual object is adjustable in size.

A method and a virtual dressing system (VDS) for rendering a representation of a user in outfits from a virtual wardrobe are provided. The VDS acquires images of outfits via scanning using a camera or from a user device to store the images in a virtual wardrobe. The VDS acquires and displays on an interactive full-length display panel weather information of predetermined locations. The VDS scans the body of the user to generate a user image and selects outfits appropriate to the acquired weather and dimensions of the generated user image from images of outfits in the virtual wardrobe or in preselected locations on a computer network. The VDS displays the images of the selected outfits. On receiving a choice of an outfit from the user, the VDS renders a representation of the user in the chosen outfit on the full-length interactive display panel, updating the rendition on request.

An augmented reality (AR) system includes a display having a 2D display coordinate system and a controller. The controller receives from an image having a 3D AR coordinate system, recognizes a first object in the image, determines a first distance to a viewpoint, and receives an instruction to display a second object image on the display, which includes a 3D AR position of the second object with a depth coordinate indicating a second distance to the viewpoint. The controller retrieves dimensions of the second object, obtains a distance index for the camera, calculates a distance scaling factor, calculates a 2D display position for the 3D AR position, calculates display dimensions for the second object, generates a display object image by scaling the second object image to the display dimensions, and displays the display object image onto the display based on a comparison of the first distance and the second distance.

Techniques described herein are directed to comparing, using a machine-trained model, neural network activations associated with data representing a simulated environment and activations associated with data representing real environment to determine whether the simulated environment is causes similar responses by the neural network, e.g., a detector. If the simulated environment and the real environment do not activate the same way (e.g., the variation between neural network activations of real and simulated data meets or exceeds a threshold), techniques described herein are directed to modifying parameters of the simulated environment to generate a modified simulated environment that more closely resembles the real environment.

The present disclosure describes techniques for face tracking. The techniques comprise receiving landmark data associated with a plurality of images indicative of at least one facial part. Representative images corresponding to the plurality of images may be generated based on the landmark data. Each representative image may depict a plurality of segments, and each segment may correspond to a region of the at least one facial part. The plurality of images and corresponding representative images may be input into a neural network to train the neural network to predict a feature associated with a subsequently received image comprising a face. An animation associated with a facial expression may be controlled based on output from the trained neural network.

A method for displaying a virtual vehicle includes: calculating a virtual world comprising the virtual vehicle and a representation of a physical object at a virtual position; calculating a virtual position of a point of view within the virtual world based on a position of the point of view at the racecourse; and calculating a portion of the virtual vehicle within the virtual world that is visible from the virtual position of the point of view, wherein the portion of the virtual vehicle visible from the virtual position of the point of view comprises a portion of the virtual vehicle that is unobscured, from the virtual position of the point of view, by the representation of the physical object at the virtual position of the physical object.

A system for rendering 6 degree-of-freedom virtual reality according to an embodiment of the present disclosure includes a visibility test module performing a visibility test for determining whether a current point of interest where a main viewpoint is directed is visible for each of a plurality of reference viewpoints and generating visibility information by identifying the number of invisible fragments of each reference viewpoint according to the performance result, a reference viewpoint selection module selecting a final reference viewpoint for a rendering process for a current frame based on the visibility information for each of the plurality of reference viewpoints and a preset selection criterion, and a rendering process module performing an image-based rendering process by using a color image and a depth image corresponding to the final reference viewpoint.

Appearance driven automatic three-dimensional (3D) modeling enables optimization of a 3D model comprising the shape and appearance of a particular 3D scene or object. Triangle meshes and shading models may be jointly optimized to match the appearance of a reference 3D model based on reference images of the reference 3D model. Compared with the reference 3D model, the optimized 3D model is a lower resolution 3D model that can be rendered in less time. More specifically, the optimized 3D model may include fewer geometric primitives compared with the reference 3D model. In contrast with the conventional inverse rendering or analysis-by-synthesis modeling tools, the shape and appearance representations of the 3D model are automatically generated that, when rendered, match the reference images. Appearance driven automatic 3D modeling has a number of uses, including appearance-preserving simplification of extremely complex assets, conversion between rendering systems, and even conversion between geometric scene representations.

A method comprises displaying in virtual reality a computer-generated scene; obtaining a movement command from a real-world physical movement of a user, the movement command corresponding to a movement of a virtual body; and adjusting the movement of the virtual body in dependence on an effect of gravity in the computer-generated scene and/or in dependence on the presence of at least one object within the computer-generated scene that inhibits the movement of the virtual body, wherein the adjusting of the movement is such that the adjusted movement of the virtual body does not correspond with the real-world physical movement of the user.