SloMo 44K: A Large-Scale Collection of Slow-Motion Videos

SloMo 44K is a large-scale collection of slow-motion video footage curated from in-the-wild sources. The dataset was created by james020619 and presented in the paper 'Seeing Fast and Slow: Learning the Flow of Time in Videos'. It was last updated on the Hugging Face platform on April 25, 2026.

Use Cases

• Training models for temporal super-resolution based on high-speed camera footage.
• Developing action recognition systems that leverage richer temporal detail.
• Researching video interpolation and frame prediction techniques using slow-motion sequences.
• Benchmarking video understanding models on footage with varied temporal dynamics.

Strengths

• Dataset is described as 'large-scale' and contains a 'collection of slow-motion video footage'.
• Videos contain 'significantly richer temporal detail than standard footage', as they are typically captured by high-speed cameras.
• Data is curated from 'in-the-wild sources', suggesting diversity in content.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and total size are unknown, which may limit suitability assessment.
• License information is unknown, which could restrict commercial or research use.

Provenance

Source

james020619 on Hugging Face.

Collection Method

Curated from in-the-wild sources, typically captured by high-speed cameras.

Freshness

Last updated 2026-04-25 02:48:50; freshness should be verified.