10-second video clips depict normal shopping and theft activities. The dataset is intended for training SlowFast action recognition models. It was sourced from the Kaggle platform, but the author, organization, and last update date are unknown.
Use Cases
- Train SlowFast models for action recognition based on video clips of shopping activities.
- Develop theft detection algorithms based on labeled video examples of theft.
- Benchmark video classification models on a domain-specific surveillance task.
- Study human activity patterns in retail environments based on video data.
Strengths
- Video clips are 10 seconds long, providing a consistent temporal unit for model training.
- The description specifies two distinct action classes: normal shopping and theft.
Limitations
- Description metadata is limited; actual data quality requires manual inspection after download.
- Row count is unknown, which may limit suitability assessment.
- Column-level documentation is absent; field semantics must be inferred after download.