Bharat Scene Text: A Novel Comprehensive Dataset and Benchmark for Indian Language Scene Text Understanding
This addresses the problem of scene text understanding for Indian languages, which is crucial for applications like assistive technology and e-commerce, but is incremental as it focuses on dataset creation and adaptation of existing methods.
The paper tackles the challenge of Indian language scene text recognition by introducing the Bharat Scene Text Dataset (BSTD), a large-scale benchmark with over 100K words across 11 Indian languages and English from 6,500 images, and shows that adapting existing English models reveals significant challenges and opportunities in this domain.
Reading scene text, that is, text appearing in images, has numerous application areas, including assistive technology, search, and e-commerce. Although scene text recognition in English has advanced significantly and is often considered nearly a solved problem, Indian language scene text recognition remains an open challenge. This is due to script diversity, non-standard fonts, and varying writing styles, and, more importantly, the lack of high-quality datasets and open-source models. To address these gaps, we introduce the Bharat Scene Text Dataset (BSTD) - a large-scale and comprehensive benchmark for studying Indian Language Scene Text Recognition. It comprises more than 100K words that span 11 Indian languages and English, sourced from over 6,500 scene images captured across various linguistic regions of India. The dataset is meticulously annotated and supports multiple scene text tasks, including: (i) Scene Text Detection, (ii) Script Identification, (iii) Cropped Word Recognition, and (iv) End-to-End Scene Text Recognition. We evaluated state-of-the-art models originally developed for English by adapting (fine-tuning) them for Indian languages. Our results highlight the challenges and opportunities in Indian language scene text recognition. We believe that this dataset represents a significant step toward advancing research in this domain. All our models and data are open source.