Awesome

Spatial Temporal Graph Convolutional Networks for Sign Language (ST-GCN-SL) Recognition

The work in this repository is described in the paper Spatial-Temporal Graph Convolutional Networks for Sign Language Recognition, available in the links below:

[ST-GCN-SL Preprint]

[ST-GCN-SL Final Paper]

This repository is a fork of the original repository available at the URL below, which is the codebase for the paper Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition Sijie Yan, Yuanjun Xiong and Dahua Lin, AAAI 2018.

[ST-GCN Github Repository]

[ST-GCN Arxiv Preprint]

ST-GCN-SL

Main project of Spatial Temporal Graph Convolutional Networks for Sign Language (ST-GCN-SL)

st-gcn/

Pre-trained weights

The pre-trained weights for the ST-GCN-SL can be downloaded through the following links.

For the training using the ASLLVD-Skeleton database:

• Weights from best epoch
• Weights from all epochs
• Explore download directory

For the training using ASLLVD-Skeleton-20 database:

• Weights from best epoch
• Weights from all epochs
• Explore download directory

ASLLVD-Skeleton-Preproc

Project containing source code for ASLLVD dataset preprocessing

asllvd-skeleton-preproc/

Log Reader

Auxiliary project for reading the ST-GCN logs

log-reader/

Contact

For any question, feel free to contact

Cleison Amorim  : cca5@cin.ufpe.br