VisDrone: Object Detection and Tracking in Drone Videos

The VisDrone dataset is a large-scale benchmark for object detection, segmentation, and tracking in drone videos. It includes a variety of challenging scenarios with diverse objects and backgrounds, converted into the YOLO format by author banu4prasad. The dataset was last updated on July 2, 2025.

Use Cases

• Training object detection models based on drone-captured images and videos.
• Benchmarking segmentation algorithms on aerial video data.
• Developing and testing tracking algorithms for objects in drone footage.
• Evaluating model performance on diverse backgrounds and challenging scenarios.

Strengths

• Dataset is described as large-scale, suggesting substantial volume.
• Includes a variety of challenging scenarios and diverse objects.
• Converted into the widely-used YOLO format for ease of integration.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file formats are unknown, which may limit suitability assessment.
• Data may reflect geographic or temporal bias inherent to the original VisDrone benchmark.

Provenance

Source

Original VisDrone benchmark dataset.

Collection Method

Converted into YOLO format.

Freshness

Last updated 2025-07-02 05:19:39.