Human motion tracking and 3D motion track detection technology based on visual information features and machine learning
Xin Zhang,Zhongqiu Xu,Hongbo Liao,Xin Zhang,Zhongqiu Xu,Hongbo Liao
DOI: https://doi.org/10.1007/s00521-021-06703-2
2022-01-07
Neural Computing and Applications
Abstract:Human motion detection is a major subject of investigation in the field of machine visualization and synthetic integration. It serves a wide and important spectrum of applications in terms of visual surveillance, cross-functional simulation, movement acquisition, and high-level man-computer interface. In order to improve the accuracy of human tracking, this paper starts from the geometric flow characteristics of the image, and proposes a Gaussian algorithm to process human motion images, and applies it to the video human motion tracking of machine learning methods. Using the statistical features of the optimized transformation as the features of the image, regression learning and prediction of the three-dimensional human body movement posture in the monocular video image. First, the optimal parameters and additional statistical features of the transformation used in the extraction of human image features are verified through experiments, and then various regression methods are used for parameter learning, and the prediction performance and human tracking tests are carried out. The final test results found that the accuracy of the method based on visual information features and machine learning is 5% higher than the previous method, and the recognition rate of human motion tracking is as high as 95%. The processing capability of the detection technology of 3D motion trajectory has also increased by nearly 7%.
computer science, artificial intelligence