Modeling Temporal Dynamics and Spatial Configurations of Actions Using Two-Stream Recurrent Neural Networks

TL;DR

This paper introduces a two-stream RNN architecture that models both temporal dynamics and spatial configurations of skeletons for action recognition, improving accuracy over traditional methods.

Contribution

The paper proposes a novel two-stream RNN framework with hierarchical and stacked structures, incorporating spatial sequence modeling and 3D data augmentation for enhanced skeleton-based action recognition.

Findings

Significant improvement on 3D action recognition benchmarks.

Effective modeling of spatial configurations improves recognition accuracy.

Data augmentation enhances model generalization.

Abstract

Recently, skeleton based action recognition gains more popularity due to cost-effective depth sensors coupled with real-time skeleton estimation algorithms. Traditional approaches based on handcrafted features are limited to represent the complexity of motion patterns. Recent methods that use Recurrent Neural Networks (RNN) to handle raw skeletons only focus on the contextual dependency in the temporal domain and neglect the spatial configurations of articulated skeletons. In this paper, we propose a novel two-stream RNN architecture to model both temporal dynamics and spatial configurations for skeleton based action recognition. We explore two different structures for the temporal stream: stacked RNN and hierarchical RNN. Hierarchical RNN is designed according to human body kinematics. We also propose two effective methods to model the spatial structure by converting the spatial graph into a sequence of joints. To improve generalization of our model, we further exploit 3D transformation based data augmentation techniques including rotation and scaling transformation to transform the 3D coordinates of skeletons during training. Experiments on 3D action recognition benchmark datasets show that our method brings a considerable improvement for a variety of actions, i.e., generic actions, interaction activities and gestures.