A 14-class Philippine vehicle detection dataset, cleaned using CleanLab. The dataset was used to train a YOLO11s model achieving 90.3% mAP50. The author, organization, and last update date are unknown.
Use Cases
- Train object detection models based on the 14 vehicle classes mentioned.
- Benchmark model performance based on the reported 90.3% mAP50 metric.
- Apply data cleaning techniques based on the described use of CleanLab.
Strengths
- Dataset has been cleaned using the CleanLab process.
- A YOLO11s model trained on this data achieved a reported 90.3% mAP50.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Data may reflect geographic bias inherent to its Philippine focus.
Provenance
- Collection Method
- Likely contains images collected for vehicle detection, cleaned with CleanLab.
- Geography
- Philippines