A U-Net and Transformer Pipeline for Multilingual Image Translation

TL;DR

This paper introduces a comprehensive multilingual image translation system combining a U-Net for text detection, Tesseract for recognition, and a custom Transformer for translation, emphasizing full customization and adaptability.

Contribution

The paper presents a novel end-to-end pipeline integrating a U-Net, Tesseract, and a from-scratch Transformer for multilingual image translation, avoiding reliance on pre-trained models.

Findings

High text detection accuracy from synthetic-trained U-Net

Effective text recognition with Tesseract on detected regions

Promising translation performance validated by BLEU scores

Abstract

This paper presents an end-to-end multilingual translation pipeline that integrates a custom U-Net for text detection, the Tesseract engine for text recognition, and a from-scratch sequence-to-sequence (Seq2Seq) Transformer for Neural Machine Translation (NMT). Our approach first utilizes a U-Net model, trained on a synthetic dataset , to accurately segment and detect text regions from an image. These detected regions are then processed by Tesseract to extract the source text. This extracted text is fed into a custom Transformer model trained from scratch on a multilingual parallel corpus spanning 5 languages. Unlike systems reliant on monolithic pre-trained models, our architecture emphasizes full customization and adaptability. The system is evaluated on its text detection accuracy, text recognition quality, and translation performance via BLEU scores. The complete pipeline demonstrates promising results, validating the viability of a custom-built system for translating text directly from images.