Investigation of Different Skeleton Features for CNN-based 3D Action Recognition
This work addresses action recognition for computer vision applications, but it is incremental as it adapts existing skeleton features to CNN methods.
The paper tackled the problem of identifying effective skeleton features for CNN-based 3D action recognition by encoding five spatial skeleton features into images and studying joint selection, achieving state-of-the-art performance with 75.32% accuracy on the NTU RGB+D dataset.
Deep learning techniques are being used in skeleton based action recognition tasks and outstanding performance has been reported. Compared with RNN based methods which tend to overemphasize temporal information, CNN-based approaches can jointly capture spatio-temporal information from texture color images encoded from skeleton sequences. There are several skeleton-based features that have proven effective in RNN-based and handcrafted-feature-based methods. However, it remains unknown whether they are suitable for CNN-based approaches. This paper proposes to encode five spatial skeleton features into images with different encoding methods. In addition, the performance implication of different joints used for feature extraction is studied. The proposed method achieved state-of-the-art performance on NTU RGB+D dataset for 3D human action analysis. An accuracy of 75.32\% was achieved in Large Scale 3D Human Activity Analysis Challenge in Depth Videos.