Performance Boosting of Scale and Rotation Invariant Human Activity Recognition (HAR) with LSTM Networks Using Low Dimensional 3D Posture Data in Egocentric Coordinates

Performance Boosting of Scale and Rotation Invariant Human Activity Recognition (HAR) with LSTM Networks Using Low Dimensional 3D Posture Data in Egocentric Coordinates

Human activity recognition (HAR) has been an active area in computer vision with a broad range of applications, such as education, security surveillance, and healthcare. HAR is a general time series classification problem. LSTMs are widely used for time series classification tasks. However, they wor...

Full description

Bibliographic Details
Main Author: Ibrahim Furkan Ince
Format: Article
Language:English
Published: MDPI AG 2020-11-01
Series:Applied Sciences
Subjects:
human activity recognition (HAR)
Kinect depth sensor
3D posture data
egocentric coordinate system
dimension reduction
discrete cosine transform (DCT)
Online Access:https://www.mdpi.com/2076-3417/10/23/8474

Internet

https://www.mdpi.com/2076-3417/10/23/8474

Similar Items