3D Human Motion Posture Tracking Method Using Multilabel Transfer Learning

Abstract

To overcome the high position and posture angle tracking error, long tracking loss time and posture tracking update response time, and low fitness problem of traditional human motion posture tracking methods, in this paper, a three-dimensional (3D) human motion posture tracking method using multilabel transfer learning is proposed. According to the human structure composition and degree of freedom constraints, the 3D human joint skeleton model is constructed to generate the 3D human pose image and perform the noise reduction operation. The background difference is used to detect the 3D human moving target. Using multilabel transfer learning, human motion posture features are extracted from joint position and joint angle, and the estimation results of 3D human motion posture are obtained. The tracking error of human motion posture is corrected by three-step search, and the visual 3D human motion posture tracking results are output. The results show that, compared with the traditional human motion posture tracking method, the position and posture angle tracking errors of the proposed method are 2.18 mm and 0.178 deg, respectively. The tracking loss time and posture tracking update response time are shorter, which proves that the proposed method has more advantages in tracking accuracy and higher adaptability.

Deep Custom Transfer Learning Models for Recognizing Human Activities via Video Surveillance

“…These include autonomous driving, surveillance, ambient-supported living, entertainment, & human-computer interaction (HCI). Two basic approaches exist for activity recognition: learning-based representations and the more traditional, manually produced feature-based representation [1]- [4]. The learning-based representation, as well as particularly, deep learning, which included a trainable feature extractor with a trainable classifier, introduced the idea of end-to-end learning first.…”