VideoEval: Action Recognition in Dark and Long Videos

A video dataset for evaluating action recognition models, compiled by author lixinhao and last updated on 2025-05-21. It includes videos from the ARID (Action Recognition in the Dark) and Breakfast (Action Recognition in Long Video) datasets. The dataset page provides links to download the original video files and annotations.

Use Cases

• Benchmarking action recognition models based on videos from the ARID dataset for dark environments.
• Training models for long-term action recognition based on videos from the Breakfast dataset.
• Evaluating model robustness across different video conditions (dark vs. standard, short vs. long) based on the combined dataset structure.

Strengths

• Links to two established source datasets (ARID and Breakfast) for action recognition.
• Dataset was updated on 2025-05-21, suggesting recent maintenance.

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, file formats, and license information are unknown, which may limit suitability assessment.

Provenance

Source

Aggregated from the ARID dataset (hosted on OpenDataLab) and the Breakfast dataset (hosted by Brown University's Serre Lab).

Collection Method

Likely compiled by the author from existing public datasets for evaluation purposes.

Freshness

Last updated 2025-05-21 02:45:49; freshness should be verified.