Action recognition based on depth image sequence

Human action recognition is the process of labeling image sequences with action labels. Robust solutions to this problem have applications in domains such as medical care, human-computer interaction and virtual training. The task is challenging for feature extraction due to variations in motion performance, recording settings and inter-personal differences. To meet these challenges, we propose two types of feature extraction methods based on the Kinect depth image sequences in this paper. One is assuming that there exists even distribute position lines in the three-dimensional space of frame difference, it will be active when the moving object touches them. The other is mapping the 16 successive frame sequences to a single image by Speed Time Mapping (STM) or Time Depth Mapping (STDM), obtaining 36-dimensiona spatial-temporal features in this image. These features are fed into Support Vector Machine (SVM) to identify the action categories. The experiments compare their performance and demonstrate the effectiveness of STDM.

[1]  Yan Song,et al.  Describing Trajectory of Surface Patch for Human Action Recognition on RGB and Depth Videos , 2015, IEEE Signal Processing Letters.

[2]  Di Wu,et al.  Recent advances in video-based human action recognition using deep learning: A review , 2017, 2017 International Joint Conference on Neural Networks (IJCNN).

[3]  François Brémond,et al.  Modeling spatial layout of features for real world scenario RGB-D action recognition , 2016, 2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS).

[4]  Tao Li,et al.  Human action recognition based on improved CoHOG-LQC , 2017, 2017 29th Chinese Control And Decision Conference (CCDC).

[5]  Albert Ali Salah,et al.  Action recognition with deep neural networks , 2017, 2017 25th Signal Processing and Communications Applications Conference (SIU).

[6]  Ahmet Burak Can,et al.  Action recognition with skeletal volume and deep learning , 2017, 2017 25th Signal Processing and Communications Applications Conference (SIU).

[7]  Chengcheng Jia,et al.  Low-Rank Tensor Subspace Learning for RGB-D Action Recognition. , 2016 .

[8]  Xiaofeng Wang,et al.  Human action recognition using transfer learning with deep representations , 2017, 2017 International Joint Conference on Neural Networks (IJCNN).

[9]  Ling Shao,et al.  Action Recognition Using 3D Histograms of Texture and A Multi-Class Boosting Classifier , 2017, IEEE Transactions on Image Processing.