VisDrone2019 Test Set Detection Performance Comparison

VisDrone2019 test set results compare detection models for drone-captured imagery. The dataset contains performance metrics in a 5.5 KB XLS file, authored by Linfeng Jia and updated in March 2026.

Use Cases

• Compare model performance metrics like mAP for small object detection tasks on drone imagery.
• Analyze detection head and feature interaction methods using reported evaluation scores.
• Benchmark Yolo11n baseline against other architectures on the VisDrone2019 test set.

Strengths

• Directly comparable results from the standardized VisDrone2019 test set.
• Compact 5.5 KB XLS file for easy access and analysis.
• Clear CC BY 4.0 license permitting reuse and redistribution.

Limitations

• Extremely limited scope at 5.5 KB, likely containing only summary statistics, not raw data.
• No raw detection outputs, bounding boxes, or image data are included for model training.
• Unknown row and column counts prevent assessment of statistical depth.

Provenance

Source

figshare, author Linfeng Jia.

Collection Method

Performance comparison on the VisDrone2019 benchmark.

Time Range

null

Freshness

Updated March 2026.

Geography

null