Skeleton-based Action Recognition via Temporal-Channel Aggregation

TL;DR

This paper introduces TCA-GCN, a novel graph convolutional network that dynamically learns and aggregates spatial and temporal features for skeleton-based action recognition, achieving superior performance on multiple datasets.

Contribution

The paper proposes a new TCA-GCN model with dynamic spatial-temporal feature aggregation and multi-scale skeletal feature fusion using attention mechanisms.

Findings

Outperforms state-of-the-art on NTU RGB+D datasets

Effective dynamic topological feature aggregation

Improved multi-scale temporal feature modeling

Abstract

Skeleton-based action recognition methods are limited by the semantic extraction of spatio-temporal skeletal maps. However, current methods have difficulty in effectively combining features from both temporal and spatial graph dimensions and tend to be thick on one side and thin on the other. In this paper, we propose a Temporal-Channel Aggregation Graph Convolutional Networks (TCA-GCN) to learn spatial and temporal topologies dynamically and efficiently aggregate topological features in different temporal and channel dimensions for skeleton-based action recognition. We use the Temporal Aggregation module to learn temporal dimensional features and the Channel Aggregation module to efficiently combine spatial dynamic channel-wise topological features with temporal dynamic topological features. In addition, we extract multi-scale skeletal features on temporal modeling and fuse them with an attention mechanism. Extensive experiments show that our model results outperform state-of-the-art methods on the NTU RGB+D, NTU RGB+D 120, and NW-UCLA datasets.