Skeleton-Based Action Recognition with Multi-Stream Adaptive Graph Convolutional Networks

TL;DR

This paper introduces MS-AAGCN, a flexible graph convolutional network with adaptive topology and multi-stream attention, significantly improving skeleton-based action recognition accuracy on large datasets.

Contribution

The work presents a novel adaptive graph construction method and a multi-stream framework with attention modules, enhancing recognition performance over fixed-topology GCNs.

Findings

Outperforms state-of-the-art on NTU-RGBD and Kinetics-Skeleton datasets.

Adaptive graph learning improves model flexibility and accuracy.

Multi-stream attention enhances focus on key joints and features.

Abstract

Graph convolutional networks (GCNs), which generalize CNNs to more generic non-Euclidean structures, have achieved remarkable performance for skeleton-based action recognition. However, there still exist several issues in the previous GCN-based models. First, the topology of the graph is set heuristically and fixed over all the model layers and input data. This may not be suitable for the hierarchy of the GCN model and the diversity of the data in action recognition tasks. Second, the second-order information of the skeleton data, i.e., the length and orientation of the bones, is rarely investigated, which is naturally more informative and discriminative for the human action recognition. In this work, we propose a novel multi-stream attention-enhanced adaptive graph convolutional neural network (MS-AAGCN) for skeleton-based action recognition. The graph topology in our model can be either uniformly or individually learned based on the input data in an end-to-end manner. This data-driven approach increases the flexibility of the model for graph construction and brings more generality to adapt to various data samples. Besides, the proposed adaptive graph convolutional layer is further enhanced by a spatial-temporal-channel attention module, which helps the model pay more attention to important joints, frames and features. Moreover, the information of both the joints and bones, together with their motion information, are simultaneously modeled in a multi-stream framework, which shows notable improvement for the recognition accuracy. Extensive experiments on the two large-scale datasets, NTU-RGBD and Kinetics-Skeleton, demonstrate that the performance of our model exceeds the state-of-the-art with a significant margin.