iSign-3D: Mediapipe Keypoints for Indian Sign Language (10% Sample)

iSign-3D is a sample dataset containing Mediapipe pose and hand keypoints for Indian Sign Language (ISL). The dataset description indicates it is a 10% sample of a larger collection. The original author, organization, and specific collection details are unknown.

Use Cases

• Train sign language recognition models based on pose and hand keypoints.
• Benchmark pose estimation algorithms on sign language data.
• Develop gesture-based interfaces using keypoint sequences.
• Analyze the kinematic structure of Indian Sign Language gestures.

Strengths

• Focuses on Indian Sign Language, a specific and less-represented language in datasets.
• Uses Mediapipe, a widely-adopted framework for pose and hand tracking.

Limitations

• Row count and dataset scale are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Collection Method

Likely extracted from video using the Mediapipe framework.

Geography

Data appears to cover Indian Sign Language.