Network algorithm real-time depth image 3D human recognition for augmented reality
- PDF / 1,507,159 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 77 Downloads / 181 Views
SPECIAL ISSUE PAPER
Network algorithm real‑time depth image 3D human recognition for augmented reality Renyong Huang1 · Mingyi Sun2 Received: 15 June 2020 / Accepted: 27 October 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract This paper studies the application of augmented reality real-time depth image technology to 3D human motion recognition technology. The accuracy and real-time performance of sensor-based 3D human reconstruction are affected by visual characteristics and illumination changes. Features are not easily extracted and cannot be tracked, leading to failures in the 3D reconstruction of the human body. Based on this system, the sensor-based visual inertial initialization algorithm is studied, which is integrated in the two-frame image time interval to provide accurate initial values for vision-based motion estimation, improve the accuracy of the calculated posture, and finally improve the accuracy of the 3D reconstruction system. Based on the relationship between the depth image and the distance and reflectivity, a model for correcting the distance error and reflectivity error of the depth image is established to improve the accuracy of the depth image, and finally the accuracy of the three-dimensional reconstruction of the human body. Keywords Real-time depth image · Augmented reality · Network algorithm · 3D human recognition
1 Introduction Augmented reality technology (AR) can fuse virtual information and real information through the recognition of real environment information and the real-time 3D rendering capability of the computer. Therefore, the virtual information expanded from the current two-dimensional screen to the real three-dimensional space environment, greatly expanding the display space of the virtual information and expanding the interactive bandwidth. In the early days of augmented reality, on the mobile terminal, research on target registration and tracking of the real environment and fusion rendering of virtual information was conducted. Therefore, most of the research on the spatial interface interaction in the AR environment is also based on the target registration technology to the real environment [1]. However, the method of target registration is pre-registered and the viewpoint of the target is not separated from that of the phone * Renyong Huang [email protected] 1
School of Literature and Journalism, Xihua University, Chengdu 610000, China
School of Computer Science, Chengdu Normal University, Chengdu 610000, China
2
camera, thus greatly limiting its interactive use scenarios [2]. Therefore, most of the research on the interaction of virtual information in the augmented reality environment aimed at scenarios with early registration goals, while there is little research on the spatial interface in the generalized environment. Therefore, the realization of the generalized recognition of augmented reality to reality information, to get rid of the dependence on target prediction, can greatly expand the use scenarios of augmented realit
Data Loading...
