Two-Stream 3D Convolutional Neural Network for Skeleton-Based Action Recognition

TL;DR

This paper introduces a novel two-stream 3D CNN model for skeleton-based action recognition, effectively capturing spatial and temporal features and outperforming RNN-based methods on benchmark datasets.

Contribution

First application of 3D CNN in skeleton-based action recognition, with a multi-temporal extension to improve global feature capturing.

Findings

Outperforms most RNN-based methods on benchmark datasets

Demonstrates robustness to noise in skeleton data

Validates the complementary nature of spatial and temporal features

Abstract

It remains a challenge to efficiently extract spatialtemporal information from skeleton sequences for 3D human action recognition. Although most recent action recognition methods are based on Recurrent Neural Networks which present outstanding performance, one of the shortcomings of these methods is the tendency to overemphasize the temporal information. Since 3D convolutional neural network(3D CNN) is a powerful tool to simultaneously learn features from both spatial and temporal dimensions through capturing the correlations between three dimensional signals, this paper proposes a novel two-stream model using 3D CNN. To our best knowledge, this is the first application of 3D CNN in skeleton-based action recognition. Our method consists of three stages. First, skeleton joints are mapped into a 3D coordinate space and then encoding the spatial and temporal information, respectively. Second, 3D CNN models are seperately adopted to extract deep features from two streams. Third, to enhance the ability of deep features to capture global relationships, we extend every stream into multitemporal version. Extensive experiments on the SmartHome dataset and the large-scale NTU RGB-D dataset demonstrate that our method outperforms most of RNN-based methods, which verify the complementary property between spatial and temporal information and the robustness to noise.