YOLOv7-Large achieves 90.4% mAP50 for ancient coin character recognition

Researchers have developed a deep learning approach using YOLO variants for character recognition on ancient Roman Republican coins. They created a new dataset of 5,654 coin images with 38,808 annotations for 21 character labels. YOLOv7-Large demonstrated the highest performance with an mAP50 of 90.4%, outperforming other YOLO versions. AI

IMPACT Demonstrates advanced object detection capabilities for specialized historical data analysis.

RANK_REASON Academic paper detailing a novel dataset and object detection approach for historical artifact analysis. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• Roman Republican denarius
• YOLOv3
• YOLOv4
• YOLOv5
• YOLOv7
• YOLOv7-Extended
• YOLOv7-Large
• YOLOv7-xl
• YOLOv8

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →