Time-frozen material shooting and collection

The system is mainly composed of camera array module and main control module, and its structure block diagram is shown in the right figure. The main control module mainly has two functions: one is to control the camera array to realize various shooting modes (such as simultaneous shooting and sequential shooting). ) and complete the automatic collection of materials. Secondly, considering the interactivity, visibility and operability of the whole system, this module enables users to complete the whole process of shooting and collecting through interface operation.

The core of the main control module is the embedded control circuit, which extends the parallel input and output interface, and the output port is used to connect the shutter circuit of the camera. The input port is used to input shooting mode selection control signals and interval time parameters during interval shooting. Considering that in some special occasions, shooting is triggered by the outside world, such as when an actor takes off, he touches a pressure sensor to send out a shooting signal, so this system also uses several input ports to receive the shooting trigger signal from the outside world, such as sensors and infrared rays. The embedded control circuit is connected with the workstation by serial level conversion circuit, which realizes two-way communication of data.

In addition to dozens of cameras fixed on the bracket, the camera array also includes a camera focusing system. To connect the images taken by multiple digital cameras into a clear and continuous video, it is required that the images taken by each camera should be relatively similar and the main object should be relatively stable at a certain position, so it is necessary to accurately locate multiple cameras.

Positioning includes two contents: calibration of camera shooting orientation (pitch angle and horizontal azimuth angle) and calibration of shooting angle (push-pull adjustment). In addition, in order to control the camera shutter and collect materials through the main control module, a shutter control unit and a data acquisition unit are also designed in the camera array, and the collected data are transmitted to the workstation through the USB interface.

The camera array in the picture uses a digital camera. The bracket is designed to be movable with multiple degrees of freedom, and the horizontal and vertical layout curves of the camera array can be arbitrarily combined according to different shooting requirements. At the same time, the bracket is equipped with a universal joint for each camera to meet the requirements of different camera positions. The camera array of the system shooting the scene is shown on the right. This is a reality show introduced from Britain, in which "time freezing" shooting technology is used.

Dream Rubik's cube recording scene

The live stage uses a cube as a closed space for players to play games. As long as the game is successful, players can get corresponding bonuses or choose a program group to help them realize their predetermined dreams. Therefore, the plan was named "Dream Cube". On the hemispherical stage outside the cube, you can see the dense camera array spread out by an arc. According to the program group, the camera actually used for shooting is a 90 SLR camera, and at the same time, a high-speed camera is deployed at both ends of the arc array to form a perfect set.

Both sides of the camera group

In program recording, the two devices will undertake different tasks respectively. An arc-shaped frame with a length of 26.9m is pulled out from the back of the cube, which almost surrounds most of the cube from the horizontal direction, and 92 points are marked on the arc-shaped frame at equal intervals (90 points in the middle are used for installing cameras, and 2 points at the head and tail are used for installing high-speed cameras). Install 9 1 SLR cameras (90 cameras form a "time freeze" system, and 1 camera is used as a backup). When shooting, the system will trigger 90 cameras to capture the actions of players in the cube.

Through field observation, it is found that the deployment mode of this system is quite high-end and complicated. On average, each camera needs at least three cables, including power cord, USB data cable and a special cable that triggers the shutter through the central control system.

In this environment and requirement, the power lines are connected layer by layer in a very strict way, which avoids the power supply failure caused by cable confusion to the greatest extent.

For image processing, because it collects a lot of information, it needs to be completed in a short time, and any microcomputer can't undertake it independently. Therefore, the topology adopted is roughly that the data lines of several cameras are connected to a USB hub, and then every few hubs are connected to a computer host, and then these hosts process the image information collected by the whole system together.

All shutter trigger wires on the camera will be connected to the trigger box uniformly.

Strict line connection

We can see that these cables have strict protection measures and clear numbers, which are very different from the shutter zone lines usually used in camera shooting.

Debugging of camera

All shooting modes are set manually.

Every time the engineer debugs, he will set a reference point in the cube and adjust the focus of all cameras to this position.

At the same time, all cameras must be set to full manual mode at the same time, and the command of shooting parameters will be handed over to manual mode. Then the staff will set the parameters of one of the cameras through the computer and synchronize with each camera in the whole cluster to ensure that the collected images will not have abnormal phenomena such as focus translocation, inconsistent brightness and depth.

The camera part is divided into two acquisition modes. Freeze-frame mode is that all cameras are triggered at the same time in an instant to obtain 90 panoramic pictures that can cover the cube. The other mode is that all cameras trigger in turn within1s. The latter mode is more commonly used in the case of the British Dream Cube, which can create a dynamic CG effect similar to that in computer games, precisely rotating 360 degrees on the same horizontal line, while the people in the picture keep moving naturally. This is achieved in the following ways.

Video workspace

It can be seen that this is a scene of division of labor and cooperation of equipment with different specifications. At the end of the lower camera group is a high-speed camera, and above it are several traditional cameras with different angles. The photographer is focusing on the monitor in front of him.

There is a high-speed camera on both sides of the camera group. The final effect of these two high-speed cameras can smoothly present an instantaneous action at a very low speed, similar to the details of the magnifying lens in sci-fi movies such as The Matrix.

You can only shoot at high speed six times per game. After the game, the director will edit the material he needs immediately and upload it to the server quickly.

For several cameras and a high-speed camera near the director's area, before each recording, the cameraman will adjust the aperture and white balance in advance according to the requirements, and the director will preview in advance according to the high-speed position of the program design, and adjust the frame rate of the computer software according to different competitions to ensure that the audience can see the TV picture with normal effect.

The whole recording process is interlocking and exciting, giving you a brand-new visual experience in static and dynamic pictures. I believe the audience will feel the mystery when the program is broadcast. Netizen Xu posted a homemade video on Tencent video. Click on the cited resources to watch.