YOLOv8m Table Extraction Model
| Property | Value |
|---|---|
| License | AGPL-3.0 |
| Framework | PyTorch (Ultralytics) |
| Accuracy | 95.19% mAP@0.5 |
| Downloads | 13,029 |
What is yolov8m-table-extraction?
YOLOv8m-table-extraction is a specialized computer vision model designed for detecting and localizing tables in documents. Built on the efficient YOLOv8 medium architecture, it can identify both bordered and borderless tables with high accuracy, making it particularly useful for document analysis and data extraction tasks.
Implementation Details
The model is implemented using the Ultralytics framework version 8.0.21 and requires ultralyticsplus for deployment. It employs advanced object detection techniques with configurable parameters for NMS confidence threshold, IoU threshold, and maximum detection limits.
- Supports detection of bordered and borderless table types
- Achieves 95.19% mAP@0.5 on validation dataset
- Implements YOLOv8 medium architecture for optimal performance
- Requires minimal setup with ultralyticsplus package
Core Capabilities
- High-precision table detection in documents
- Flexible configuration options for detection parameters
- Support for batch processing of images
- Easy integration with existing document processing pipelines
Frequently Asked Questions
Q: What makes this model unique?
This model specializes in table detection with the ability to distinguish between bordered and borderless tables, achieving high accuracy while maintaining the efficiency of the YOLOv8 architecture. Its high download count (13,029) demonstrates its practical utility in real-world applications.
Q: What are the recommended use cases?
The model is ideal for document processing systems, automated data extraction from scanned documents, digital document analysis, and any application requiring accurate table detection in various document formats. It's particularly useful in scenarios where both traditional bordered tables and modern borderless designs need to be detected.
