Synthetic CCTV Fall Detection Dataset with Human Pose Keypoints

An open-source synthetic dataset for computer vision tasks, created by Simuletic. It is designed for fall detection, pose estimation, and incident monitoring from overhead CCTV camera perspectives. The dataset includes keypoint annotations for human posture, which may help models distinguish between standing and fallen individuals.

Use Cases

• Training fall detection models based on annotated human posture keypoints.
• Developing pose estimation algorithms for overhead camera perspectives.
• Benchmarking incident monitoring systems using synthetic CCTV footage.

Strengths

• Includes keypoint (pose) annotations, which are not present in standard object detection datasets.
• Specifically designed for overhead CCTV perspectives, a common deployment scenario.

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 is unknown, which may limit suitability assessment.

Provenance

Source

Simuletic

Collection Method

Synthetically generated data.

Freshness

Last updated 2025-12-14 18:37:06; freshness should be verified.