An optical motion capture system consists of an infrared optical lens, motion capture software, reflective marking points, POE switches, cables, calibration frames, and lens fixtures such as tripods.
First of all, the infrared optical lens through the tripod, clamps and other lens fixtures arranged around the site to ensure that the lens field of view can cover the capture area, and then all the lenses through the network cable connected to the POE switch. The lenses are powered and data transferred through the POE switch and connected to the motion capture software in the computer. After the software is started, first connect the motion capture lenses in real-time mode operation in the page.
After the system hardware and software set up and connected to each other successfully, the next step is the site of the calibration, the role of the calibration is to establish the XYZ coordinate system for the motion capture area, calculate the position and attitude of each lens in the coordinate system, only after the completion of the calibration, you can correctly get to the site of the three-dimensional coordinates of each Marker point of the data, which is divided into L-type calibration and T-type calibration.
L-type calibration by placing the L-type calibration rod in the center of the field, the corresponding settings in the software to complete, its purpose is twofold: first, to determine a unified coordinate system, through the positioning of the four points on the L-type rod, the system can distinguish between its long and short axes, so as to define the orientation of the world's coordinate axes and the location of the origin, and, secondly, the process can give the lens to see the L-type rod an initial parameter, as the initial value of the later parameter optimization. as the initial value for later parameter optimization.
The role of T-calibration is to give each lens enough data to be able to carry out an iterative parameter optimization based on the original initial value. In this process, the software is in T-calibration mode, the operator holds the T-bar and swings it around the field, and the lens captures a large amount of data in real time.
After completing the calibration, the spatial data of the captured object can be acquired. In the need to locate people or objects on the surface of the reflective marking points (a surface coated with a special reflective material of the silver gray ball), the motion capture lens on the LED lights to the outside of the launch of infrared light, and at the same time to receive the reflective marking points reflected back to the infrared light. When more than one optical lens at the same time "see" a marking point, this marking point in the space of the three-dimensional position will be determined.
The reflective marker position information obtained by the lens needs to be transferred to the computer in real time for data processing and use. In the optical motion capture system, all the lenses are connected to the switch through the network cable, when the lens to obtain the spatial position of the reflective marking point information, this information will be transmitted through the network cable to the switch, and then the switch unified transmission to the computer connected to the real-time motion capture software to receive.
After the software obtains the three-dimensional spatial position of multiple reflective markers, the next step is to recognize the object. In the same object surface affixed with multiple reflective markers, the distance between these specific points is unchanged, thus, the same object affixed with the point of naming, and will be connected to the line between the points, indicating that the interrelationship between the two points, this set of point names and connecting line information in the software is recorded as a set of Markerset. when the object with this set of Markerset information appears in the When an object with this set of Markerset information appears in the field, it is recognized by the system as an independent object.
Some human motion capture requires a large amount of sticker capture data, and there are specialized sticker models for this purpose. According to the provided sticker models, reflective markers are pasted on the human body at fixed locations, and the points are identified, connected, and bound to the bones in the software.
When the system is able to recognize the captured object in real time, a complete optical motion capture system has been established, and then can directly carry out motion capture, the model data obtained from the capture can also be adjusted and corrected in real time according to the effect in the software. According to the needs of different fields, the optical motion capture system can also be synchronized with the force platform and other equipment for motion and force data capture, connected to the 3D software for virtual character generation and other operations.