stephen-cheng / table_detection_with_detectron2_n_mask_rcnn

A detectron2 based method and powered by the PyTorch deep learning framework for table detection.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Install required packages:

pip install -r requirements.txt

Install Detectron2

python -m pip install 'git+https://github.com/facebookresearch/detectron2.git'

Create custom COCO dataset

Then you can run the voc2coco.py script to generate a COCO data formatted JSON file for you.

python voc2coco.py ./dataset/annotations ./dataset/coco/output.json

Further instruction on how to create your own datasets, read the THIS.

Then you can run the following Jupyter notebook to visualize the coco annotations.

COCO_Image_Viewer.ipynb

Training

python table_detect_train.py

Evaluation

python table_detect_test.py

The result looks like this:

About

A detectron2 based method and powered by the PyTorch deep learning framework for table detection.


Languages

Language:Jupyter Notebook 92.9%Language:Python 7.1%