Dual Action Graph

Independent Dual Graph Attention Convolutional Network for skeleton-based action recognition Representation modeling learning with multi-domain decoupling for unsupervised skeleton-based action recognition Multi-scale sampling attention graph convolutional networks for skeleton-based action recognition

In skeleton-based action recognition, graph convolutional networks GCNs, which model the human body skeletons as spatiotemporal graphs, have achieved remarkable performance. However, in existing GCN-based methods, the topology of the graph is set manually, and it is fixed over all layers and input samples. This may not be optimal for the hierarchical GCN and diverse samples in action

In our proposed Hybrid Dual-branch Network HDBN, we use robust skeleton modality for human action recognition. Conceptually, the sequence of skeletal structure forms a nat-ural topological graph, wherein joints represent graph ver-tices and bones serve as edges. The graph can be defined as G V,L, where V v 1,v 2,,v Nand L l 1

The crucial issue in current methods for skeleton-based action recognition how to comprehensively capture the evolving features of global context information and temporal dynamics, and how to extract discriminative representations from skeleton joints and body parts. To address these issues, this article proposes a dual-excitation spatial-temporal graph convolutional method. The method

Keywords Action recognition Skeleton Graph convolutional networks Dual-domain Spatial-temporal Spectral 1 Introduction Action recognition is a challenging task in the eld of computer vision. And it is at the forefront of applications to understand the human social activity Islam and Iqbal 2020.

Skeleton-based action recognition through dual-granularity feature fusion with self-adapting graph convolution and multi-scale temporal convolution. we supplemented one newly proposed methods, DualHead-Net 37, a dual-head graph network in this experiment. The experimental results of MTGCN and seven competitors were listed in Table 2.

The red graph is the dual graph of the blue graph, and vice versa.. In the mathematical discipline of graph theory, the dual graph of a planar graph G is a graph that has a vertex for each face of G.The dual graph has an edge for each pair of faces in G that are separated from each other by an edge, and a self-loop when the same face appears on both sides of an edge.

In this work, we propose a dual-stream structured graph convolution network DS-SGCN to solve the skeleton-based action recognition problem.The spatio-temporal coordinates and appearance contexts of the skeletal joints are jointly integrated into the graph convolution learning process on both the video and skeleton modalities.

Work process of graph convolution networks for skeleton-based action recognition. The graph convolution networks use the adjacency matrix to convolve the skeleton graph of the current frame and then convolve the state of each skeleton point in different frames. the novel independent dual graph attention convolution mechanism of IDGAN has

This section introduces two sorts of graph convolution operations according to graph signal processing GSP for skeleton action recognition. 3.1 Vertex-domain graph convolution. GCNs have been a widely used architecture since the work of Yan et al. .By constructing the skeleton data into graph 92GV, E92 with N joints and T frames, a vertex-domain graph convolution operation is defined