CompTLL-UNet: Compressed Domain Text-Line Localization in Challenging Handwritten Documents using Deep Feature Learning from JPEG Coefficients

TL;DR

This paper introduces CompTLL-UNet, a deep learning model that localizes text lines directly from JPEG compressed images, achieving state-of-the-art results while reducing storage and computational costs.

Contribution

It presents a novel approach that performs text-line localization directly in the JPEG compressed domain using a modified U-Net architecture, avoiding full decompression.

Findings

Achieves state-of-the-art performance on benchmark datasets

Reduces storage and computational costs

Effective in handling complex handwritten document issues

Abstract

Automatic localization of text-lines in handwritten documents is still an open and challenging research problem. Various writing issues such as uneven spacing between the lines, oscillating and touching text, and the presence of skew become much more challenging when the case of complex handwritten document images are considered for segmentation directly in their respective compressed representation. This is because, the conventional way of processing compressed documents is through decompression, but here in this paper, we propose an idea that employs deep feature learning directly from the JPEG compressed coefficients without full decompression to accomplish text-line localization in the JPEG compressed domain. A modified U-Net architecture known as Compressed Text-Line Localization Network (CompTLL-UNet) is designed to accomplish it. The model is trained and tested with JPEG compressed version of benchmark datasets including ICDAR2017 (cBAD) and ICDAR2019 (cBAD), reporting the state-of-the-art performance with reduced storage and computational costs in the JPEG compressed domain.