we provide a simple universal spatial modeling method perpendicular to the RNN model enhancement. Specifically, we select a set of simple geometric features, motivated by the evolution of previous work. With experiments on a 3-layer LSTM framework, we observe that the geometric relational features based on distances between joints and selected lines outperform other features and achieve state-of-art results on four datasets.
If any part of our paper and code is helpful to your work, please generously cite with:
@InProceedings{Zhang_2017_geometric,
author = {Zhang, Songyang and Liu, Xiaoming and Xiao, Jun},
title = {On Geometric Features for Skeleton-Based Action Recognition Using Multilayer LSTM Networks},
booktitle = {WACV},
year = {2017}
}