HandX: High-Fidelity Bimanual Hand Motion Data with Text Annotations

HandX provides high-fidelity bimanual hand motion data, MANO-format hand parameters, and fine-grained text annotations. This dataset accompanies the research paper 'HandX: Scaling Bimanual Motion and Interaction Generation' and is hosted by author alexzhang598. The dataset page was last updated on June 1, 2026.

Use Cases

• Train models for bimanual hand motion generation based on the provided high-fidelity motion data.
• Develop text-to-motion systems using the fine-grained text annotations.
• Research hand pose estimation and parameterization using the included MANO-format hand parameters.

Strengths

• Data includes high-fidelity bimanual hand motion sequences.
• Provides MANO-format hand parameters, a standard for 3D hand modeling.
• Includes fine-grained text annotations for motion sequences.

Limitations

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

Provenance

Source

huggingface

Collection Method

Released to accompany the HandX research paper.

Freshness

Last updated 2026-06-01 20:20:35; freshness should be verified.