Synthetic ATM Robbery Detection Video Frames with Bounding Boxes

Synthetic video frames for training object detection models to identify robbery threats at ATMs. The dataset likely contains images annotated with bounding boxes for guns and knives. It is hosted on Kaggle, but details on the creator, size, and update date are unknown.

Use Cases

• Training object detection models for security applications based on the described bounding box annotations.
• Benchmarking model performance on detecting weapons like guns and knives in CCTV footage.
• Developing synthetic data pipelines for rare event scenarios like ATM robberies.

Strengths

• Focuses on a specific, high-impact use case for automated security.
• Annotations for multiple threat types (guns and knives) are mentioned, which could support multi-class detection.

Limitations

• Row count and dataset scale are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The synthetic nature of the data may not fully capture real-world CCTV variability.

Provenance

Source

Kaggle

Collection Method

Synthetically generated, as indicated by the description.

Time Range

null

Freshness

Last update date is unknown; freshness unverified.

Geography

null