arXiv Computer Vision August 31, 2026 By Tathagata Ghosh, Sai Madhusudan Gunda, Simran Singh Sandral, Ravi Kiran Sarvadevabhatla

UniLipi: A Unified Multi-Script OCR for Historical Indic Manuscripts

Read the original on arXiv Computer Vision →

UniLipi is a unified multi‑script OCR model trained on 13 Indic scripts to recognize handwritten manuscripts under challenging conditions such as varied line geometry, length, and interruptions by non‑textual elements. It uses script‑aware synthetic data generation to perform well even with limited real annotated data. The model also predicts script identity and per‑line character counts, aiding manuscript cataloging, and its representations transfer to contemporary Indic handwriting and several non‑Indic scripts.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

computer-vision

arXiv Machine Learning

Aug 28

Systematic Literature Review of Machine Learning Models and Applications for Text Recognition

This systematic literature review examines 97 studies on optical character recognition (OCR) from 2015 to 2025, tracing the evolution of AI models, application domains, data types, and linguistic coverage. It identifies key OCR models, evaluates their performance, strengths, and limitations, and highlights unresolved challenges such as limited resources for underrepresented languages, high variability in handwritten text, and constraints in real‑time applications. The review proposes promising approaches—including self‑supervised learning, multimodal AI, AutoML, AI‑assisted postprocessing, TinyML, and joint corpora creation—to enhance OCR accuracy and address these challenges for industrial use.

By Nuzhat Khan, Ab Al-Hadi Ab Rahman, Shahriyar Masud Rizvi, Ibrahim Yousef Alshareef, Muhammad Nadzir Marsono, Muhammad Paend Bakht, Mohd Shahrizal Rusli, Shahidatul Sadiah

computer-vision multimodal

arXiv AI

Aug 11

TongGuOCR: A Layout-Aware and Token-Augmented OCR Framework for Chinese Historical Documents

arXiv:2608. 07917v1 Announce Type: new Abstract: Chinese historical documents preserve valuable cultural heritage, but many collections remain accessible only as scanned page images, preventing full-text retrieval, collation, and computational analysis.

By Zhongheng Zhou, Yi Sun, Huiguo He, Yuyi Zhang, Peirong Zhang, Yulin Fang, Dezhi Peng, Minghui Liao, Lianwen Jin

llms computer-vision multimodal benchmarks

arXiv Computer Vision

3d ago

MuSTEL: A Large-Scale Multilingual Synthetic Text-Line Image Dataset for OCR

arXiv:2610.34148v1 Announce Type: new Abstract: Optical character recognition (OCR) systems depend strongly on the scale and diversity of their training data. One approach to document OCR first analy...

By Majid Adibian, Hossein Saberi, Hossein Zeinali

computer-vision

arXiv AI

Aug 12

TongGuOCR: A Layout-Aware and Token-Augmented OCR MLLM for Chinese Historical Documents

arXiv:2608. 07917v2 Announce Type: replace Abstract: Chinese historical documents preserve valuable cultural heritage, but many collections remain accessible only as scanned page images, preventing full-text retrieval, collation, and computational analysis.

By Zhongheng Zhou, Yi Sun, Huiguo He, Yuyi Zhang, Peirong Zhang, Yulin Fang, Dezhi Peng, Minghui Liao, Lianwen Jin

llms computer-vision multimodal benchmarks

Hugging Face Trending Papers

Aug 19

Impact of Iterative Fine-Tuning on Transcription Accuracy in Complex Historical Sanskrit Manuscripts

The paper presents a local traditional OCR pipeline that can be iteratively fine‑tuned on both layout and appearance levels of complex historical Sanskrit manuscripts. By adapting to the specific manuscript distribution, the pipeline improves transcription accuracy across subsequent pages, reducing the need for costly human annotation. The authors apply the method to three manuscripts, release a dataset with detailed layout and Unicode annotations in PAGE‑XML format, and benchmark the results against leading multimodal large language models.

llms computer-vision fine-tuning multimodal benchmarks

arXiv Computer Vision

Sep 2

A Text Recognition Dataset from Sahidic Coptic Ancient Manuscripts

The paper introduces SCAM, a line-level dataset of digitized Sahidic Coptic ancient manuscripts designed for Handwritten Text Recognition in low-resource settings. SCAM captures realistic challenges such as varied acquisition conditions, ink fading, bleed-through, and material deterioration, while also presenting linguistic difficulties due to the scarce resources, uncommon alphabet, and dialect-specific diacritics of Sahidic Coptic. The authors benchmark several state‑of‑the‑art HTR methods, demonstrating the performance gap between modern, well‑resourced scripts and historically grounded, low‑resource scenarios.

By Fabio Quattrini, Carmine Zaccagnino, Costanza Bianchi, Silvia Cascianelli, Rita Cucchiara

benchmarks