This paper contributes to the challenge of skeleton-based human action recognition in videos. The key step is to develop a generic network architecture to extract discriminative features for the spatio-temporal skeleton data. In this paper, we propose a novel module, namely Logsig-RNN, which is the combination of the log-signature layer and recurrent type neural networks (RNNs). The former one comes from the mathematically principled technology of signatures and log-signatures as representations for streamed data, which can manage high sample rate streams, non-uniform sampling and time series of variable length. It serves as an enhancement of the recurrent layer, which can be conveniently plugged into neural networks. Besides we propose two path transformation layers to significantly reduce path dimension while retaining the essential information fed into the Logsig-RNN module. Finally, numerical results demonstrate that replacing the RNN module by the Logsig-RNN module in SOTA networks consistently improves the performance on both Chalearn gesture data and NTU RGB+D 120 action data in terms of accuracy and robustness. In particular, we achieve the state-of-the-art accuracy on Chalearn2013 gesture data by combining simple path transformation layers with the Logsig-RNN. Codes are available at \url{https://github.com/steveliao93/GCN_LogsigRNN}.
翻译:本文有助于应对视频中基于骨架的人类行动识别挑战。 关键步骤是开发一个通用网络架构, 用于为spatio- 时空骨质数据提取歧视性特征。 在本文中, 我们提出一个新模块, 即 Logsig- RNNN, 即日志签名层和常规型神经网络的组合。 前者来自签名和日志签名的数学原则技术, 以显示流数据, 它可以管理高样本率流、 非统一抽样和时间序列的变长。 它可以增强常态层, 并方便地插入神经网络。 除了我们提议两个路径转换层, 以大幅降低路径尺寸, 同时保留输入Logsig- RNNNNNN 模块的基本信息。 最后, 数字结果显示, SOTA 网络的Losig- RNNNM 模块取代 RNNNM 模块, 持续提高CO- RTU/ D120 行动数据的性能性, 准确性和稳健性。 特别是, 我们通过Star- f State- R- NG- mainal- gillingsalbislationsal_ calbs kills.