metadata
license: apache-2.0
Suryolo : Layout Model For Arabic Documents
Suryolo is combination of Surya layout Model form SuryaOCR(based on Segformer) and YoloV10 objection detection.
Setup Instructions
Clone the Surya OCR GitHub Repository
git clone https://github.com/VikParuchuri/surya.git
cd surya
Switch to v0.4.14
git checkout f7c6c04
Install Dependencies
The author has not provided requirements.txt file, but environment.yml from our conda environment has been uploaded, This file can be used to recreate environment for Suryolo model.
Suryolo Pipeline
Download surya_yolo_pipeline_copy.cpython-310-x86_64-linux-gnu.so , yolov10x_best.pt and surya folder from the Repository.
Place surya_yolo_pipeline_copy.cpython-310-x86_64-linux-gnu.so, yolov10x_best.pt and surya folder in same directory (They are dependent on each other).
from surya_yolo_pipeline_copy import suryolo # This import will originate from surya_yolo_pipeline_copy.cpython-310-x86_64-linux-gnu.so , which is present in the repo. Also this works with Linux based OS only.
from surya.postprocessing.heatmap import draw_bboxes_on_image
image_path = "sample.jpg"
image = Image.open(image_path)
bboxes = suryolo(image_path)
plotted_image = draw_bboxes_on_image(bboxes,image)