Sony Pictures Entertainment Inc.

A multi-camera rig apparatus including: a central ring including a plurality of couplers spaced about the central ring; a plurality of rails and modular attachments, each rail and modular attachment coupled to each of the plurality of couplers; a moon ring assembly coupled to each rail and modular attachment, the moon ring assembly including a semi-circular rail and an indexing mechanism configured to receive a spring-loaded steel ball; and a camera plate including a first mounting surface and a second mounting surface, the first mounting surface attached to the moon ring assembly and the second mounting surface configured for mounting to a camera.

Video capture of a subject, including: a first IR camera, a second IR camera, and a color camera, for capturing video data of the subject; a post, where the first IR camera, the second IR camera, and the color camera are attached to the post, and where the color camera is positioned between the first IR camera and the second IR camera; at least one IR light source, for illuminating the subject; and a processor configured to: generate depth solve data for the subject using data from the first IR camera and the second IR camera; generate projected color data by using data from the color camera to project color onto the depth solve data; and generate final capture data by merging the depth solve data and the projected color data.

Reducing data used during capture in a physical capture volume by selectively activating image capture devices from a virtual view, including: setting up a virtual camera to receive information about and visualize the physical capture volume and a plurality of image capture devices in the virtual view; providing, to the virtual camera, the virtual view of the physical capture volume with a capability to move around the physical capture volume and activate or deactivate each of the plurality of image capture devices; calculating a view frustum, wherein the view frustum is a region of 3-D space within the physical capture volume that would appear on a view screen of the virtual camera; and defining the view frustum of the virtual camera which intersects with the plurality of image capture devices defined in the virtual view.

Systems, methods, and apparatus to generate training data include receiving a beam of light reflected from a subject and splitting the beam of light into at least a first beam and a second beam; directing the first beam toward a first sensor and the second beam toward a second sensor; sensing and processing the first beam to generate a first video having a first characteristic; sensing and processing the second beam to generate a second video having a second characteristic; and processing the first video and the second video to generate training data.

Methods, systems, and computer-readable storage medium including: a lens and a focal reducer to receive a beam of image; a beam splitter to receive the beam of image from the focal reducer and split the beam of image into multiple directions. The system also includes a plurality of sensors coupled to the beam splitter. Each sensor of the plurality of sensors is configured to sense the beam of image within a particular band of frequencies. Further, the particular band of frequencies of a first sensor of the plurality of sensors does not overlap with the particular band of frequencies of a second sensor of the plurality of sensors. The focal reducer condenses and amplifies the beam of image to increase the sensitivity of each sensor.

Calibrating cameras in a rig, including: projecting light patterns onto surfaces of objects in a capture volume; capturing image data of the light patterns projected onto the surfaces of the objects using the cameras; and processing the captured image data to estimate a calibration data of the cameras, wherein processing the image data to estimate the calibration data includes measuring distortions in the light patterns.

Reducing data used during capture in a physical capture volume by selectively activating image capture devices from a virtual view, including: setting up a virtual camera to receive information about and visualize the physical capture volume and a plurality of image capture devices in the virtual view; providing, to the virtual camera, the virtual view of the physical capture volume with a capability to move around the physical capture volume and activate or deactivate each of the plurality of image capture devices; calculating a view frustum, wherein the view frustum is a region of 3-D space within the physical capture volume that would appear on a view screen of the virtual camera; and defining the view frustum of the virtual camera which intersects with the plurality of image capture devices defined in the virtual view.

Machine vision and control including: instructing a plurality of imaging devices to capture or detect images to calibrate the plurality of imaging devices; recognizing the captured or detected images using machine vision or image recognition; determining a configuration of the plurality of imaging devices using the recognized images; and adjusting, positioning, aligning, and calibrating the plurality of imaging devices using the determined configuration.

Methods, systems, and computer-readable storage medium including: a lens and a focal reducer to receive a beam of image; a beam splitter to receive the beam of image from the focal reducer and split the beam of image into multiple directions. The system also includes a plurality of sensors coupled to the beam splitter. Each sensor of the plurality of sensors is configured to sense the beam of image within a particular band of frequencies. Further, the particular band of frequencies of a first sensor of the plurality of sensors does not overlap with the particular band of frequencies of a second sensor of the plurality of sensors. The focal reducer condenses and amplifies the beam of image to increase the sensitivity of each sensor.

Capturing data from a camera rig including a plurality of sensors to capture data during a capture period, including: a plurality of image sensors including at least a visible light sensor and an infrared (IR) sensor, the plurality of image sensors configured to have a common center line through a single lens; a plurality of light sources including a visible light source and an IR source, wherein the IR source projects an IR beam of a calibration pattern; and a processor coupled to the plurality of image sensors and the plurality of light sources, wherein the processor is configured to process data captured by the visible light sensor, and wherein the processor is configured to process data captured by the IR sensor.