External Validation Results for DSE-YOLO11 on BDD100K

DSE-YOLO11, a lightweight object detector, improved recall from 0.744 to 0.811 and mAP50 from 0.810 to 0.856 on the RAD dataset. This 5.5 KB Excel file contains external validation results for the model on the BDD100K dataset, authored by Yange Chen and last updated in June 2026. The data provides preliminary performance metrics for detecting traffic elements in complex road scenes.

Use Cases

• Benchmarking lightweight object detectors based on reported recall and mAP50 metrics.
• Validating model performance on external datasets based on the BDD100K evaluation.
• Analyzing detection robustness for small and irregular targets in complex road scenes as described.

Strengths

• Performance metrics (recall, mAP50) are explicitly quantified in the description.
• The dataset is small (5.5 KB), facilitating quick inspection and sharing.
• Released under a permissive CC-BY-4.0 license for open use.

Limitations

• The dataset is extremely small (5.5 KB), suggesting very limited scope, likely summary statistics only.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Yange Chen

Collection Method

Likely contains experimental results from validating the DSE-YOLO11 model.

Freshness

Last updated 2026-06-01 17:34:32; freshness should be verified.