This repository contains a paper collection of the methods for document image processing, including appearance enhancement, deshadowing, dewarping, deblurring, binarization and so on.
DIAR software for synthetic document image and groundtruth generation, with various degradation models for data augmentation
a pure C# implementation of GCDRNet (Appearance Enhancement for Camera-captured Document Images in the Wild, IEEE TAI 2023) — document-image shadow removal and appearance enhancement without onnxruntime or any other third-party inference engine. It ships its own compiled-model inference engine
Contains code for object detection models like RetinaNet, FasterRCNN, YOLO that can be used to detect and recognize tables in document images.
Realistic synthetic degradation of document images — simulate scans, phone photos, photocopies, and aged paper for data augmentation in document AI / OCR / layout analysis pipelines. 23 physically-motivated transforms, 6 presets, CLI + Python API.
To associate your repository with the document-image topic, visit your repo's landing page and select "manage topics."