Hand Landmark Data for Sign Language Gestures

AlphaSign is a dataset of hand gestures with extracted landmarks formatted for Graph Neural Networks. The dataset is designed for sign language recognition tasks. The author, organization, and specific volume are not provided.

Use Cases

• Train GCN models to classify sign language letters from hand landmark graphs.
• Benchmark landmark-based gesture recognition algorithms against pre-processed graph data.
• Develop real-time sign language interpreters using structured hand pose features.

Strengths

• Data is pre-processed and structured for immediate use with Graph Neural Networks.
• Focuses on the specific application domain of sign language recognition.

Limitations

• The total number of gesture samples, rows, or participants is unknown.
• The dataset's geographic and demographic diversity is unspecified, which may limit model generalizability.

Provenance

Source

Kaggle

Collection Method

Hand gestures were likely recorded and processed to extract landmark coordinates, though the specific capture method is not detailed.