|
53 | 53 |
|
54 | 54 | Surya-Tabled 使用内置ocr模块,表格模型为行列识别模型,无法识别单元格合并,导致分数较低 |
55 | 55 |
|
56 | | -| 方法 | TEDS | TEDS-only-structure | |
57 | | -|:------------------------------------------------------------------------------------------------|:-----------:|:-------------------:| |
58 | | -| [surya-tabled(--skip-detect)](https://github.com/VikParuchuri/tabled) | 0.33437 | 0.65865 | |
59 | | -| [surya-tabled](https://github.com/VikParuchuri/tabled) | 0.33940 | 0.67103 | |
60 | | -| [deepdoctection(rag-flow)](https://github.com/deepdoctection/deepdoctection?tab=readme-ov-file) | 0.59975 | 0.69918 | |
61 | | -| [ppstructure_table_master](https://github.com/PaddlePaddle/PaddleOCR/tree/main/ppstructure) | 0.61606 | 0.73892 | |
62 | | -| [ppsturcture_table_engine](https://github.com/PaddlePaddle/PaddleOCR/tree/main/ppstructure) | 0.67924 | 0.78653 | |
63 | | -| [StructEqTable](https://github.com/UniModal4Reasoning/StructEqTable-Deploy) | 0.67310 | 0.81210 | |
64 | | -| [RapidTable(SLANet)](https://github.com/RapidAI/RapidTable) | 0.71654 | 0.81067 | |
65 | | -| table_cls + wired_table_rec v1 + lineless_table_rec | 0.75288 | 0.82574 | |
66 | | -| table_cls + wired_table_rec v2 + lineless_table_rec | 0.77676 | 0.84580 | |
67 | | -| [RapidTable(SLANet-plus)](https://github.com/RapidAI/RapidTable) | **0.84481** | **0.91369** | |
| 56 | +| 方法 | TEDS | TEDS-only-structure | |
| 57 | +|:---------------------------------------------------------------------------------------------------------|:-----------:|:-------------------:| |
| 58 | +| [surya-tabled(--skip-detect)](https://github.com/VikParuchuri/tabled) | 0.33437 | 0.65865 | |
| 59 | +| [surya-tabled](https://github.com/VikParuchuri/tabled) | 0.33940 | 0.67103 | |
| 60 | +| [deepdoctection(table-transformer)](https://github.com/deepdoctection/deepdoctection?tab=readme-ov-file) | 0.59975 | 0.69918 | |
| 61 | +| [ppstructure_table_master](https://github.com/PaddlePaddle/PaddleOCR/tree/main/ppstructure) | 0.61606 | 0.73892 | |
| 62 | +| [ppsturcture_table_engine](https://github.com/PaddlePaddle/PaddleOCR/tree/main/ppstructure) | 0.67924 | 0.78653 | |
| 63 | +| [StructEqTable](https://github.com/UniModal4Reasoning/StructEqTable-Deploy) | 0.67310 | 0.81210 | |
| 64 | +| [RapidTable(SLANet)](https://github.com/RapidAI/RapidTable) | 0.71654 | 0.81067 | |
| 65 | +| table_cls + wired_table_rec v1 + lineless_table_rec | 0.75288 | 0.82574 | |
| 66 | +| table_cls + wired_table_rec v2 + lineless_table_rec | 0.77676 | 0.84580 | |
| 67 | +| [RapidTable(SLANet-plus)](https://github.com/RapidAI/RapidTable) | **0.84481** | **0.91369** | |
68 | 68 |
|
69 | 69 | ### 使用建议 |
70 | 70 | wired_table_rec_v2(有线表格精度最高): 通用场景有线表格(论文,杂志,期刊, 收据,单据,账单) |
|
0 commit comments