Spatial Temporal Graph Convolutional Networks (ST-GCN) for Skeleton-Based Action Recognition in PyTorch Resources

Spatial Temporal Graph Convolutional Networks (ST-GCN) for Skeleton-Based Action Recognition in PyTorch Resources

该项目整合了 RTMPose 姿态识别和 ST-GCN 时空图卷积两个模型，结合开创性的距离角度判定算法成功实现了“基于人体姿态关键点的视频分类和打分任务”。

st-gcn是aaai 2018提出的经典的基于骨骼的行为识别模型，通过将图卷积应用在具有拓扑结构的人体骨骼数据上，使用时空图卷积提取时空特征进行行为识别，极大地提升了基于骨骼的行为识别任务精度。

ST-GCN is able to exploit local pattern and correlation from human skeletons. Below figures show the neural response magnitude of each node in the last layer of our ST-GCN.

pytorch gcn tcn gnn road-traffic-prediction Resources. Readme License. LGPL-2.1 license Activity. Stars. 545 stars. Watchers. 3 watching. Forks. 116 forks. Report repository Releases. No releases published. Packages 0. No packages published . Contributors 3. hazdzz Chieh Chang; jeongwhanchoi Jeongwhan Choi; ATRM-Raphael;

ST-GCN is able to exploit local pattern and correlation from human skeletons. Below figures show the neural response magnitude of each node in the last layer of our ST-GCN.

ST-GCN is able to exploit local pattern and correlation from human skeletons. Below figures show the neural response magnitude of each node in the last layer of our ST-GCN.

skeleton-based action recognition (ST-GCN) 2D pose estimation; skeleton-based action generation; 3D pose estimation; pose tracking; build custom skeleton-based dataset; create …

The framework STGCN consists of two spatio-temporal convolutional blocks (ST-Conv blocks) and a fully-connected output layer in the end. Each ST-Conv block contains two temporal gated convolution layers and one spatial graph convolution layer in the middle.