RoMo-SMPLX: Single-Person Body Motion Sequences with Text Descriptions

RoMo-SMPLX is a large-scale dataset of single-person body motion sequences in the SMPL-X parameter space, paired with rich multi-level text descriptions. It is the raw representation of the RoMo body corpus, with the same motions also released in other feature formats. The dataset is authored by RoMoDataset and was last updated on May 21, 2026.

Use Cases

• Train text-to-motion generation models based on the paired multi-level text descriptions.
• Analyze human body pose and movement patterns based on SMPL-X parameter sequences.
• Benchmark motion feature extraction algorithms using the raw SMPL-X representation.
• Develop cross-modal retrieval systems linking natural language to body motion.

Strengths

• Large-scale collection of single-person body motion sequences.
• Pairs motion data with rich multi-level text descriptions.
• Provides raw SMPL-X parameter representation, offering flexibility for feature extraction.

Limitations

• Dataset is currently private/internal; public release date is unspecified.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file size are unknown, which may limit suitability assessment.

Provenance

Source

RoMoDataset

Freshness

Last updated 2026-05-21 17:56:06; freshness should be verified.