Detr TableDetection 5000 Images
D
Detr TableDetection 5000 Images
Developed by illuin
A table detection model fine-tuned based on facebook/detr-resnet-50, trained on the table_detection_light dataset
Downloads 247
Release Time : 2/2/2023
Model Overview
This model is used to detect table regions in images, implemented based on the DETR (Detection Transformer) architecture
Model Features
Transformer-based Object Detection
Adopts the DETR architecture, avoiding complex anchor box designs and non-maximum suppression steps in traditional object detection methods
Lightweight Training Data
Fine-tuned with only 5000 images, suitable for resource-constrained application scenarios
End-to-End Training
Directly outputs detection results without complex post-processing steps
Model Capabilities
Table detection in images
Table region localization
Use Cases
Document Processing
PDF Document Analysis
Detect table regions from scanned PDF documents
Office Automation
Automatically identify table locations in documents for subsequent processing
Featured Recommended AI Models
Š 2025AIbase