Multi-view stereoscopic vision system and method for indoor scene virtual reality live broadcast

Abstract

The invention discloses a multi-view stereoscopic vision system for indoor scene virtual reality live broadcast. The multi-view stereoscopic vision system for indoor scene virtual reality live broadcast includes M indoor guide rails, N independent cameras sliding on the M guide rails, N drive dollies which can drive the cameras to move on the guide rails, a processor and virtual reality output equipment, wherein M>=1, and N>=3; the N independent cameras can be combined two by two to shoot a target object, and can transmit the shot information to the processor; after processing the information, the processor transmits virtual reality video signals to the virtual reality output equipment; and a user can observe the target object through the virtual reality output equipment. As the plurality of cameras can form multi pairs of binocular stereoscopic vision systems two by two, the multi-view stereoscopic vision system and system for indoor scene virtual reality live broadcast can observe and shoot the object to guarantee that the environment information of the object and the surrounding environment of the object can be maximally extracted so as to enable 3D modeling to be more exquisite.

Description

technical field [0001] The invention relates to the field of virtual reality applications, in particular to a multi-eye stereo vision system and method applied to indoor scene virtual reality live broadcasting. Background technique [0002] Nowadays, with the continuous development of virtual reality technology, the camera system based on the principle of binocular vision has played an increasingly important role in the industrial production and real life of human society. Different from the general traditional monocular camera method, binocular vision uses two cameras to shoot the same object, observes an object from two or more points, and obtains images under different perspectives, according to the pixel matching between images The relationship between pixels is calculated by the principle of triangulation to obtain the three-dimensional information of the object. After obtaining the depth of field information of the object, the actual distance between the object and th...

AI Technical Summary

Problems solved by technology

[0005] In summary, although the binocular stereo vision system can measure the depth of field information of the target object, it cannot provide 360° all-round shooting due to the small coverage of the two cameras.

The multi-eye vision system that uses a pair of main-eye cameras and multiple auxiliary-eye cameras also loses the flexibility of the multi-eye vision system because the main-eye camera and the auxiliary-eye cameras are fixed on the same component. The range is limited by the movement of the main eye camera, making it impossible to capture the target object from multiple angles at the same time

Images