Q. What frame rates do your currently available synchronous camera array systems support?
A. 24, 25, 30, 50, 60, 100, 120 and 240fps.
Q. How many cameras are used?
A. We recommend at least 96 cameras for one or two performer full-body motion photogrammetry-based captures.
Q. What is the resolution of the cameras?
A. 5472 x 3648 (5.5K / 20 megapixels) at 24fps. 1920 x 1080 (2K / 2 megapixels) at other frame rates.
Q. Is there a maximum recording time per take?
A. At 5.5K resolution the currently available portable 96 camera system is limited by the cameras' internal image buffers to 8.9 seconds per take (214 frames at 24fps). At 2K resolution the current system can record for several minutes or longer at up to 120fps. Our camera control systems also support cameras with larger buffers and higher overall resolutions and frame rates (i.e. 4K 120fps with no time limit).
Q. What is the size of the rig - and the size of the capture area within the rig?
A. Our rigs are designed to be both portable and scalable. The smallest footprint is 4m in diameter with a 1.5m wide x 2m high capture area. They then scale up in 1m steps to 8m in diameter with a 6m wide x 3m high capture area. In addition to the rigs themselves being scalable the cameras have motorized Zeiss zoom lenses with which we can change the area of interest from full body to close-up in smaller rig configurations, or we can change the size of the capture area in larger rig configurations, allowing a greater or lesser amount of movement of the subject in exchange for capture resolution. Zoom settings, like all camera settings, can be changed en masse within a few seconds between takes.
Q. Do you provide the lighting?
A. Yes. We provide 97 CRI LED flicker-free constant illumination flex lighting that mounts to the rig and moves with the vertical elements of the rig when the rig changes sizes. The total output of the LED lighting is 4,000 watts, which is approximately equivalent to 16K of tungsten lighting.
Q. How much power is required?
A. 18-24 amps for the lighting. 6-10 amps for the camera system. 2-24 amps for on-set processing. Power can be provided entirely through multiple 220v 50 Hz single phase outlets - or, in 60 Hz regions, through multiple 110v 60 Hz single phase outlets. Alternatively, a single 400v 63 amp 50 Hz three phase CEE (3P+N+E) outlet socket (3x 63A, 5wire) with IP44 coupling can be provided in 50 Hz regions.
Q. Were projectors, strobes, infrared lasers, or infrared cameras used in the production of the examples on this page?
A. No. Our systems do, however, provide sync signals for all of the above if needed.
Q. Is green screen required?
A. No. Green screen is not necessary for pure photogrammetry-based processes, nor was it used in the above example. The rig and background are excluded from photogrammetric reconstruction based on depth. Green screen can be used for the removal of unwanted items within the capture area such as the floor, rigs, treadmills, props, garments, limbs, face, head, hair, etc.
Q. Is the system silent and does it record audio?
A. Yes. Silent and fanless. The cameras record at precise sync sound speeds for dual-system sync sound, timecode and multitrack audio recording.
Q. What output formats are supported?
A. 2D image sequences, 3D .obj sequences, and 4D Alembic sequences. Other output formats on request.
Q. What is the resolution and file size of the .obj files?
A. The full-body example in the demo video above was solved at an average of 500,000 faces per frame with 4K textures resulting in an average lightly compressed file size of 60 megabytes per frame (mesh plus texture). This file size for each 3D frame is approximately equivalent to the size of a single uncompressed 20 megapixel still image (24 bit 8 bits per channel RGB). Higher or lower resolution meshes and textures can be produced by adjusting the photogrammetry parameters.
Q. How much data is generated?
A. The bandwidth of the compressed .obj sequence above (meshes plus textures) is approximately 1.5 gigabytes per second. The bandwidth of the original image data recorded by the cameras is approximately 13 gigabytes per second (20,544 10x compressed 20 megapixel 24 bit 8 bits per channel jpegs, ~5.5 megabytes per image at 2,304 images per second). Compression and resolution settings are variable and using different compression and resolution settings will result in different file sizes.
Q. Can we take delivery of the original camera images and process them ourselves - or take them to another vendor for processing?
A. Of course. In fact we recommend doing so. Multiple third-party software solutions for photogrammetry are available. Depending on the purpose of your assets you may wish to reprocess the original images with different levels of compression or into additional formats as needed.
Q. Are there standards for 4D media?
A. Yes and no. For motion picture visual effects the unofficial industry standard is .obj sequences (which is what we deliver). There are also a number of non-standard proprietary compression formats for streamable and cacheable 4D content. Methods for converting unstructured live-action 4D performances into structured animations that can be used as playable assets in game engines are underway. Open standards for all of the above are needed. If you would like to contribute to the development of open standards please contact us.
Q. How are your systems so portable when other systems with similar specs are fixed in large facilities?
A. As with all of our camera systems our 4D systems are designed specifically to be transported anywhere in the world: to where the actors are for visual effects shots and for location-based live events.
Q. Do we need to provide miscellaneous equipment insurance for shoots outside of your studio, and if so, how much insurance is required?
A. Yes. The insurable value for the 96 camera system is under 250,000 Swiss francs - similar to our other large camera arrays.
Q. Can these systems also shoot fixed-path virtual camera movement shots if the cameras are arranged in a linear array?
A. Yes. They are very fast in single shot mode (single shots download in less than two seconds), and their ability to capture bursts of 5.5K images at up to 24fps can be used to get important shots in one take. Also, as a result of their high frame rates and versatility when functioning as still cameras they can be used to produce almost every variation imaginable of traditional camera array visual effects. You can find more information about our fixed-path camera array configurations and techniques on the services page.
Q. Do you also sell these systems?
A. Yes. Our newest camera control systems are available for select pilot 4D studio customers through cametric.com.
For qualified collectors, the massive 120 camera Canon film camera rig used on The Matrix (1999) and the 80 camera 4D Sony digital camera rig used used on Ghost in the Shell (2017) are available for sale or lease to collections outside of the US. Several of our original Timetrack™ film cameras from the 1990s, the prototype of which is in the permanent collection of the Smithsonian National Museum of American History, are also available to qualified museums worldwide.
A prospective that visually details our experience and expertise providing interactive installations worldwide, with links to the media produced, can be downloaded here.