Uma breve introdução à tecnologia de imagem 3D

Com o uso de módulos de transmissão e recepção infravermelhos, 3D imaging technology can realize the collection of depth datas（such as position and details） of the shooting object and then restore the real scene. Atualmente, there is three types of 3D imaging technology: (1)Structured Light: Typical companies include Apple(who acquired the biggest structure light company Prime Sense), Microsoft(Kinect-1), Intel(RealSense), Google(Project Tango) etc. The majority of startups follow this technical route now. (2)Binocular Stereo Vision: Representative company is Leap Motion. (3)ToF: Typical company is Microsoft(Kinect-2).

Structured light is based on the principle of optical triangle measurement. According to the degree of distortion of light spot, it can measure object’s far and near angles. Then the camera will collect changes of light spot datas and then reconstructed the object by algorithm. For example, when it is a direct light, the size of light spot reflects the distance of object; while it is in a slanting light, the focal length of elliptical light spot reflects the angle between the object and lens’ section.

Binocular Stereo Vision is an important form of machine vision. Based on the parallax principle and using imaging equipments to obtain two images of the object from different positions, we can obtain three-dimensional geometric information of the object through calculating the position deviation between the corresponding points of the image.

ToF stands for Time of Flight, which obtains the exact distance of object through sending laser pulse to the target continously and then receiving the reflected light with a sensor. By detecting the flight light of the laser pulse, we can get the exact distance of the target object.