Abstract: Multiple detection techniques are used to determine whether the individual pixels in an image are background or foreground. The outputs from the multiple techniques are combined according to a confidence metric associated with each technique, which may vary from pixel to pixel. The resulting blended value for each pixel is compared with a threshold to determine whether the pixel is foreground or background.

Abstract: A camera is mounted on a drone, which is programmed to follow a flight path that is specified by trajectory parameters. The position of the drone is coordinated with a subject to be filmed by the camera, and may be coordinated with the position of one or more objects, including other automatically controlled drones, a manually controlled drone and virtual assets. The drone can also be coordinated with the behavior of a subject.

Abstract: A movie set is divided into at least a directing station and a set station, remote from each other. Actors are either in their own remote acting station or, if the director is also the cameraman, in the directing station. At the set station, a digital record of a real or virtual scene is created. Actors are recorded separately from the set and composited into the scene as they are acting. A real camera films the actor, who is automatically composited into the scene. A virtual camera films the actor and the virtual scene. Cameramen filming the actors wear virtual reality headsets, so they inhabit the virtual or pre-recorded real world into which the actors are composited, in real time. Virtual viewfinders are displayed in the headsets to show the image that the camera is capturing.

Abstract: An immersive computing machine makes a recording of an immersive MR scene. The recording includes sufficient data to reconstruct multiple different three-dimensional views of the scene depending on a perspective of a viewer using another immersive computing machine. One or multiple viewers may enter a playback of the recorded immersive scene and make further immersive recordings. The scene may be a still or a motion scene. The immersive recording is transmitted with a two-dimensional version of the recording that can be viewed on a non-immersive device, which can add the three-dimensional immersive recording to a queue for later viewing on an immersive device.

Abstract: Multiple detection techniques are used to determine whether the individual pixels in an image are background or foreground. The outputs from the multiple techniques are combined according to a confidence metric associated with each technique, which may vary from pixel to pixel. The resulting blended value for each pixel is compared with a threshold to determine whether the pixel is foreground or background.

Abstract: An unmounted camera, which is used to capture images of a user of a virtual reality application, wirelessly transmits its position, orientation and the camera feed to a computer. The computer determines the location and orientation of the camera in the virtual world and renders a view of the virtual scene from the perspective of the camera. The computer compensates for latency in the camera feed. The user's background is removed from the camera feed and the image of the user is composited with the rendered scene to result in a mixed reality scene. The composited scene is displayed back to the camera operator, either locally or on a separate screen. The system provides the freedom to the camera operator to introduce camera movement in mixed reality video productions.

Abstract: A camera is mounted on a drone, which is programmed to follow a flight path that is specified by trajectory parameters. The position of the drone is coordinated with a subject to be filmed by the camera, and may be coordinated with the position of one or more objects, including other automatically controlled drones, a manually controlled drone and virtual assets. The drone can also be coordinated with the behavior of a subject.

Abstract: A movie set is divided into at least a directing station and a set station, remote from each other. Actors are either in their own remote acting station or, if the director is also the cameraman, in the directing station. At the set station, a digital record of a real or virtual scene is created. Actors are recorded separately from the set and composited into the scene as they are acting. A real camera films the actor, who is automatically composited into the scene. A virtual camera films the actor and the virtual scene. Cameramen filming the actors wear virtual reality headsets, so they inhabit the virtual or pre-recorded real world into which the actors are composited, in real time. Virtual viewfinders are displayed in the headsets to show the image that the camera is capturing.