A U-Net and Transformer Pipeline for Multilingual Image Translation
This addresses the problem of translating text directly from images for users needing multilingual support, but it is incremental as it integrates existing components with custom training.
The paper tackles multilingual image translation by developing a custom pipeline that combines a U-Net for text detection, Tesseract for text recognition, and a from-scratch Transformer for translation, achieving promising results validated through BLEU scores.
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.