Comments (5)
公开出来的版本在后处理的时候确实没有使用gpma的分割结果。主要是考虑到:
论文中也提到gpma的分割结果只是用来辅助判断“空白单元格”是否应合并或分割开,但是在实际的表格识别场景中,空单元格的分开与否对于表格理解的意义不是很大,整体对于评测最终的性能提升也比较有限,在推理阶段加入这部分反而会额外增加一些计算开销。但是这个模块在训练阶段的加入可以使模型收敛的效果更好,因此在训练阶段是加入了全局分割的分支。
如果想看gpma的预测mask效果,可以在推理的代码中进行相应的修改即可。
from davar-lab-ocr.
@qiaoliang6 感谢回复。明白,我理解是英文的空白单元格其实没啥用,但是中文文档中空白单元格其实还挺重要的,特别是对于某一列只有空白单元格的情况。如果我想利用gpma的mask来 生成空白单元格的bbox大概改哪里可以呢
from davar-lab-ocr.
@qiaoliang6 感谢回复。明白,我理解是英文的空白单元格其实没啥用,但是中文文档中空白单元格其实还挺重要的,特别是对于某一列只有空白单元格的情况。如果我想利用gpma的mask来 生成空白单元格的bbox大概改哪里可以呢
在lgpma.py的主流程文件lgpma.py中, L162-L167中是有将分割结果打包一起存放在results中的,在后处理的时候把这部分取出来使用就可以。
不过我个人感觉单凭分割学习到的结果其实准确率提升有限,因为空格子里面没有明显的像素差异,而且标注起来都会很有歧义。我觉得反而可以试试一些基于虚拟表格线的方法。
from davar-lab-ocr.
@qiaoliang6 我把gpma的mask写出去看了确实在空白单元格的地方有一点但是较弱。另外能具体讲下虚拟表格线的做法和标注吗,我搜了下没找到相关参考。
from davar-lab-ocr.
@qiaoliang6 我把gpma的mask写出去看了确实在空白单元格的地方有一点但是较弱。另外能具体讲下虚拟表格线的做法和标注吗,我搜了下没找到相关参考。
from davar-lab-ocr.
Related Issues (20)
- how to i use inference?
- train with 1 gpu HOT 1
- 生成的表格
- 18
- how to convert LGPMA to onnx HOT 2
- 2 dead links on LPGMA page HOT 1
- DI 数据集使用问题,可视化"labels",在纵向阅读顺序多列情况遇到的问题
- The pretrained model cannot be downloaded HOT 2
- 请问CTUNet为何要将ComFintab的中文部分和英文部分分别训练
- Can CTUNet opensource the inference code without gt_bbox、gt_text
- Does CTUNet 's masked self-attention mechanism in the graph attention network USE the neighbours relation in Structural Graph Construction part? HOT 1
- In order to reproduce the accuracy in the LGPMA paper, is it necessary to turn on refine_bboxes during post-processing?
- 关于LGPMA训练数据格式的疑问
- Insufficient space for training VSR
- MMdistributedDataparallel 多卡加载数据报错,单卡没有问题
- 请问DRCC工作会开源吗
- LGPMA 空单元格坐标不能预测出来?没找到相关代码
- LGPMA表格处理
- 。
- when i run bash test.sh command i have error
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from davar-lab-ocr.