FineGym-skeleton: 2D Human Pose Sequences for Gymnastic Action Recognition

The Gym288-skeleton dataset is a human skeleton-based action recognition benchmark derived from the FineGym dataset. It provides temporally precise, fine-grained annotations of gymnastic actions along with 2D human pose sequences extracted from original video frames. The dataset was created by Lozumi and was last updated on Hugging Face in May 2026.

Use Cases

• Fine-grained action recognition based on 2D human pose sequences.
• Modeling temporally corrupted or incomplete actions based on skeleton data.
• Benchmarking skeleton-based action recognition models on gymnastic movements.

Strengths

• Derived from the Gym288 subset of the FineGym dataset, a known source for gymnastic actions.
• Provides temporally precise, fine-grained annotations for actions.
• Includes 2D human pose sequences extracted from video frames.

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

Derived from the Gym288 subset of the FineGym dataset.

Collection Method

2D human pose sequences were extracted from original video frames.

Freshness

Last updated 2026-05-21 13:03:46; freshness should be verified.