Scale and Rotation Invariant Human Activity Recognition based on Body Relative Direction in Egocentric Coordinates
| Main Authors: | Sheilla Wesonga, Ibrahim Furkan Ince, Jang Sik Park |
|---|---|
| Format: | Proceeding Journal |
| Bahasa: | eng |
| Terbitan: |
, 2020
|
| Subjects: | |
| Online Access: |
https://zenodo.org/record/4474917 |
Daftar Isi:
- Human activity recognition is widely used in various fields, such as, surveillance, education, and healthcare. In this paper, a scale and rotation invariant human activity recognition system is presented. Kinect depth sensor is employed to obtain skeleton joints. Instead of using joint angles, the angle of each limb with X, Y, Z axes of the proposed local coordinate system are employed as feature vector. Since angles are used, proposed system is already scale invariant. In order to provide rotation invariance, body relative direction in egocentric coordinates is calculated. The 3D vector between right-hip and left-hip is used to get the horizontal axis and its cross-product with the vertical axis of global coordinate system is assumed to be the depth axis of the proposed local coordinate system. As the system parameters, 8 number of limbs and their corresponding angles with X, Y, Z axes of the proposed coordinate system are employed as the feature vector. Finally, extracted features are trained and tested with LSTM (Long Short-Term Memory) network which is an artificial recurrent neural network (RNN) architecture. Experimental results indicate that proposed framework achieves outstanding results with 98.2 accuracy in average.