Method for realizing three-dimensional reconstruction of movement by using binocular camera

Abstract

The invention relates to a method for realizing three-dimensional reconstruction of movement by using a binocular camera. In a traditional method, a sensor is utilized to realize capture, while a mechanical system has the shortcoming that the captured movement authenticity is still dependent on the ability and perseverance of an animator to a great extent, and an optical system for capturing is expensive in equipment cost, and is only suitable for the application of manufacturing film television with large investment. The method for realizing three-dimensional reconstruction of movement by using the binocular camera comprises the following steps: (1) achieving the image detection of the binocular camera; (2) achieving the image capture of the binocular camera; (3) collecting multiple chessboard pictures for shooting; (4) calibrating the camera by a camera calibration tool box self-provided by Matlab; (5) completing the stereoscopic imaging of the binocular camera; (6) obtaining three-dimensional environment information by using the reprojectImageTo3D function of Open CV; (7) attaching luminous balls on major joints of a human body. The method provided by the invention is mature, can obtain a better reconstruction effect, and is superior to other vision-based three-dimensional reconstruction methods.

Description

technical field [0001] The invention belongs to the field of three-dimensional motion reconstruction based on video, in particular to a method for realizing three-dimensional motion reconstruction by using a binocular camera. Background technique [0002] At present, there are many methods to realize the three-dimensional reconstruction and capture of human body motion, and the traditional method is to use sensors to capture, mainly including mechanical systems, which are composed of potentiometers that measure the positions and directions of human joint points. The disadvantages are The realism of captured motion still largely depends on the ability and perseverance of the animator, for example: the use of electromagnetic systems for capture, which is currently a more popular motion capture system, which uses sensors that measure electromagnetic fields to capture The position data and angle data of the joint points of the real object. The disadvantage of this method is that...

AI Technical Summary

Problems solved by technology

[0002] At present, there are many methods to realize the three-dimensional reconstruction and capture of human body motion, and the traditional method is to use sensors to capture, mainly including mechanical systems, which are composed of potentiometers that measure the positions and directions of human joint points. The disadvantages are The realism of captured motion still largely depends on the ability and perseverance of the animator, for example: the use of electromagnetic systems for capture, which is currently a more popular motion capture system, which uses sensors that measure electromagnetic fields to capture The position data and angle data of the joint points of the real object. The disadvantage of this method is that it is very sensitive to capturing metal objects in the scene and there are restrictions on the reconstruction of the movement that cannot be too complicated, and this equipment is very expensive; in addition, the optical system is used to capture, This type of system attaches sensors to the human body that needs to record movement. First, a special camera is used to shoot a video with high contrast, and then the human body movement is extracted from the video. Generally, the system requires 4-6 cameras. This technology can provide high Motion data sampling rate, but the same equipment is expensive, only suitable for some high-input film and television production and other applications, not suitable for software development of personal computers

Images