Corpus ID: 236428765

Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition

@article{Chen2021ChannelwiseTR,
  title={Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition},
  author={Yuxin Chen and Ziqi Zhang and Chunfeng Yuan and Bing Li and Ying Deng and Weiming Hu},
  journal={ArXiv},
  year={2021},
  volume={abs/2107.12213}
}
Graph convolutional networks (GCNs) have been widely used and achieved remarkable results in skeleton-based action recognition. In GCNs, graph topology dominates feature aggregation and therefore is the key to extracting representative features. In this work, we propose a novel Channel-wise Topology Refinement Graph Convolution (CTR-GC) to dynamically learn different topologies and effectively aggregate joint features in different channels for skeleton-based action recognition. The proposed CTR… Expand

References

SHOWING 1-10 OF 37 REFERENCES
Dynamic GCN: Context-enriched Topology Learning for Skeleton-based Action Recognition
TLDR
This paper proposes Dynamic GCN, in which a novel convolutional neural network named Context-encoding Network (CeN) is introduced to learn skeleton topology automatically, and achieves state-of-the-art performance on three large-scale benchmarks, namely NTU-RGB+D, NTU+D 120 and Skeleton-Kinetics. Expand
Spatio-Temporal Inception Graph Convolutional Networks for Skeleton-Based Action Recognition
TLDR
This work designs a simple and highly modularized graph convolutional network architecture for skeleton-based action recognition by repeating a building block that aggregates multi-granularity information from both the spatial and temporal paths. Expand
Decoupling GCN with DropGraph Module for Skeleton-Based Action Recognition
TLDR
This paper rethink the spatial aggregation in existing GCN-based skeleton action recognition methods and discovers that they are limited by coupling aggregation mechanism, and proposes decoupling GCN to boost the graph modeling ability with no extra computation, no extra latency, noextra GPU memory cost, and less than 10% extra parameters. Expand
Skeleton-Based Action Recognition With Shift Graph Convolutional Network
TLDR
The proposed Shift-GCN notably exceeds the state-of-the-art methods with more than 10 times less computational complexity, and is composed of novel shift graph operations and lightweight point-wise convolutions. Expand
Actional-Structural Graph Convolutional Networks for Skeleton-Based Action Recognition
TLDR
The proposed AS-GCN achieves consistently large improvement compared to the state-of-the-art methods and shows promising results for future pose prediction. Expand
An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognition
TLDR
A novel Attention Enhanced Graph Convolutional LSTM Network (AGC-LSTM) for human action recognition from skeleton data can not only capture discriminative features in spatial configuration and temporal dynamics but also explore the co-occurrence relationship between spatial and temporal domains. Expand
Stronger, Faster and More Explainable: A Graph Convolutional Baseline for Skeleton-based Action Recognition
TLDR
This work proposes an efficient but strong baseline based on Graph Convolutional Network (GCN), where three main improvements are aggregated, i.e., early fused Multiple Input Branches (MIB), Residual GCN (ResGCN) with bottleneck structure and Part-wise Attention (PartAtt) block. Expand
Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation
TLDR
This paper introduces a global spatial aggregation scheme, which is able to learn superior joint co-occurrence features over local aggregation, and consistently outperforms other state-of-the-arts on action recognition and detection benchmarks like NTU RGB+D, SBU Kinect Interaction and PKU-MMD. Expand
Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition
TLDR
A novel model of dynamic skeletons called Spatial-Temporal Graph Convolutional Networks (ST-GCN), which moves beyond the limitations of previous methods by automatically learning both the spatial and temporal patterns from data. Expand
Skeleton-Based Action Recognition With Directed Graph Neural Networks
TLDR
A novel directed graph neural network is designed specially to extract the information of joints, bones and their relations and make prediction based on the extracted features and is tested on two large-scale datasets, NTU-RGBD and Skeleton-Kinetics, and exceeds state-of-the-art performance on both of them. Expand
...
1
2
3
4
...