摘要:
A system and method executable by a computing device of an augmented reality system for registering a camera in a physical space is provided. The method may include identifying an origin marker in a series of images of a physical space captured by a camera of an augmented reality system, and defining a marker graph having an origin marker node. The method may further include analyzing in real-time the series of images to identify a plurality of expansion markers with locations defined relative to previously imaged markers, and defining corresponding expansion marker nodes in the marker graph. The method may further include calculating a current position of the camera of the augmented reality system in the physical space based on a location of a node in the marker graph corresponding to a most recently imaged marker, relative to the origin marker and any intermediate markers.
摘要:
Systems and methods for mesh generation from depth images are provided. According to one aspect, a method executable by a compression device for sending compressed depth information is provided. The method may comprise, at a compression module executed on the compression device, receiving a depth image of a scene from a depth camera. The depth image may include a matrix of pixels, each pixel in the matrix including a depth value indicating a depth of an object in the scene observed at that pixel. The method may further comprise compressing the depth image into a tree data structure, and sending the tree data structure via a communication path to a rendering device for generating a mesh of the scene at the rendering device.
摘要:
Systems and methods for mesh generation from depth images are provided. According to one aspect, a method executable by a compression device for sending compressed depth information is provided. The method may comprise, at a compression module executed on the compression device, receiving a depth image of a scene from a depth camera. The depth image may include a matrix of pixels, each pixel in the matrix including a depth value indicating a depth of an object in the scene observed at that pixel. The method may further comprise compressing the depth image into a tree data structure, and sending the tree data structure via a communication path to a rendering device for generating a mesh of the scene at the rendering device.
摘要:
A system and method executable by a computing device of an augmented reality system for registering a camera in a physical space is provided. The method may include identifying an origin marker in a series of images of a physical space captured by a camera of an augmented reality system, and defining a marker graph having an origin marker node. The method may further include analyzing in real-time the series of images to identify a plurality of expansion markers with locations defined relative to previously imaged markers, and defining corresponding expansion marker nodes in the marker graph. The method may further include calculating a current position of the camera of the augmented reality system in the physical space based on a location of a node in the marker graph corresponding to a most recently imaged marker, relative to the origin marker and any intermediate markers.
摘要:
In a virtual reality or mixed reality environment, an HMD device is configured to use surface reconstruction data points obtained with a sensor package to identify a location of a floor of a real world environment in which the device operates by sorting the data points by height into respective buckets where each bucket holds a different range of heights. A bucket having the greatest number of data points that are below the height of a user of the HMD device is used to identify the height of the real world floor, for example, by calculating an average of height values of data points in that bucket. A floor for the virtual reality environment may then be aligned to the identified height of the real world floor.
摘要:
An HMD device renders a virtual reality environment in which areas of the real world are masked out so that real world objects such as computer monitors, doors, people, faces, and the like appear visible to the device user and no holographic or virtual reality content is rendered over the visible objects. The HMD device includes a sensor package to support application of surface reconstruction techniques to dynamically detect edges and surfaces of the real world objects and keep objects visible on the display as the user changes position or head pose or when the real world objects move or their positions are changed. The HMD device can expose controls to enable the user to select which real world objects are visible in the virtual reality environment.
摘要:
A mixed-reality head mounted display (HMD) device supports a three dimensional (3D) virtual world application with which a real world desktop displayed on a monitor coupled to a personal computer (PC) may interact and share mouse input. A mouse input server executing on the PC tracks mouse movements on the desktop displayed on a monitor. When movement of the mouse takes it beyond the edge of the monitor screen, the mouse input server takes control of the mouse and stops mouse messages from propagating through the PC's system. The mouse input server communicates over a network connection to a mouse input client exposed by the application to inform the client that the mouse has transitioned to operating in the virtual world and passes mouse messages describing movements and control operation such as button presses.
摘要:
An HMD device is configured to check the placement of newly introduced objects in a virtual reality environment such as interactive elements like menus, widgets, and notifications to confirm that the objects are significantly present within the user's field of view. If the intended original placement would locate the object outside the field of view, the HMD device relocates the object so that a portion of the object is viewable at the edge of the HMD display closest to its original placement. Such smart placement of virtual objects enables the user to readily discover new objects when they are introduced into the virtual reality environment, and then interact with the objects within a range of motions and/or head positions that is comfortable to support a more optimal interaction and user experience.
摘要:
A head mounted display (HMD) device operating in a real world physical environment is configured with a sensor package that enables determination of an intersection of a device user's projected gaze with a location in a virtual reality environment so that virtual objects can be placed into the environment with high precision. Surface reconstruction of the physical environment can be applied using data from the sensor package to determine the user's view position in the virtual world. A gaze ray originating from the view position is projected outward and a cursor or similar indicator is rendered on the HMD display at the ray's closest intersection with the virtual world such as a virtual object, floor/ground, etc. In response to user input, such as a gesture, voice interaction, or control manipulation, a virtual object is placed at the point of intersection between the projected gaze ray and the virtual reality environment.
摘要:
A head mounted display (HMD) device operating in a real world physical environment is configured with a sensor package that enables determination of an intersection of a projection of the device user's gaze with a location in a mixed or virtual reality environment. When a projected gaze ray is visibly rendered on other HMD devices (where all the devices are operatively coupled), users of those devices can see what the user is looking at in the environment. In multi-user settings, each HMD device user can see each other's projected gaze rays which can facilitate collaboration in a commonly-shared and experienced mixed or virtual reality environment. The gaze projection can be used much like a finger to point at an object, or to indicate a location on a surface with precision and accuracy.