ScanTailor Advanced is the version that merges the features of the ScanTailor Featured and ScanTailor Enhanced versions, brings new ones and fixes.

[ICML 2024] BiLLM: Pushing the Limit of Post-Training Quantization for LLMs

This repository contains a paper collection of the methods for document image processing, including appearance enhancement, deshadow, dewarping, deblur, and binarization.

[CVPR 2020] This project is the PyTorch implementation of our accepted CVPR 2020 paper : forward and backward information retention for accurate binary neural networks.

Compressive Autoencoder.

A Local Adaptive Thresholding framework for image binarization written in C++, with JS and Python bindings. Implementing: Otsu, Bernsen, Niblack, Sauvola, Wolf, Gatos, NICK, Su, T.R. Singh, WAN, ISauvola, Bataineh, Chan and Shafait.

Orchestra is a sheet music reader (optical music recognition (OMR) system) that converts sheet music to a machine-readable version.

This project is the official implementation of our accepted ICLR 2022 paper BiBERT: Accurate Fully Binarized BERT.

Document Image Binarization

Some recent Quantizing techniques on PyTorch

[ICML 2023] This project is the official implementation of our accepted ICML 2023 paper BiBench: Benchmarking and Analyzing Network Binarization.

ShabbyPages is a state-of-the-art corpus of born-digital document images with both ground truth and distorted versions appropriate for use in training models to reverse distortions and recover to original denoised documents.

This repository contains source code to binarize any real-value word embeddings into binary vectors.

Improving Document Binarization via Adversarial Noise-Texture Augmentation (ICIP 2019)

[NeurIPS 2023] This project is the official implementation of our accepted NeurIPS 2023 paper BiMatting: Efficient Video Matting via Binarization.

Pytorch implementation of BiFSMNv2, TNNLS 2023

Orchestra is a sheet music reader (optical music recognition (OMR) system) that converts sheet music to a machine-readable version.

This is a jupyter notebook with 8 different solutions for common problems of digital image processing, including object recognition and binarization using adaptative threshold.

Maximum entropy named-entity recognition (NER)

Old book pages (with groundtruth), formerly used for OCR studies. There are several versions of the set (concerning resolution and binarization). Noised and denoised sets (done by several methods) are eventually going to be uploaded.