metadata
base_model:
- PaddlePaddle/RT-DETR-L_wireless_table_cell_det
RT-DETR Fine-Tuned Table Cell Detection Model
This model is a fine-tuned RT-DETR-L_wireless_table_cell_det model for detecting individual table cells in images.
It is based on the Real-Time DEtection TRansformer architecture and is optimized for precise table cell boundary detection.
Fine-Tuning Details
- Base model:
RT-DETR-L_wireless_table_cell_det - Task: Table cell boundary detection
- Data type: Image
- File format: TXT
- Training data: 947,642 examples
- Data size: 40.1 GB
- Epochs: 20
Evaluation
| Metric | Before | After |
|---|---|---|
| Precision | 0.718 | 0.770 |
Government R&D Project
This software is a result of the following government research and development project:
| Item | Details |
|---|---|
| Project Name | AI ๊ธฐ๋ฐ ์ค์๊ธฐ์ ์ ๋ฌด ์๋ํ ๋ฐ ํตํฉ ๊ฒ์ ์๋ฃจ์ |
| Project No. | RQT-25-090153 |
| Supervisor | ์ ๋ณดํต์ ์ฐ์ ์งํฅ์ |
| Performer | Proten Co., Ltd. |
| R&D Period | 2026.01.01 ~ 2026.06.30 |