Mudit Gupta

Papers

Total Citations

H-Index

About

Mudit Gupta is a researcher at the intersection of natural language processing and computer vision, with a particular focus on signboard transliteration using deep learning. His work addresses the critical challenge of converting multilingual text from real-world signage into machine-readable, transliterated formats—a task essential for navigation, accessibility, and language technology in diverse linguistic environments. Gupta’s most-cited paper, "Review of Signboard Transliteration Using Deep Learning" (2021), provides a comprehensive survey of deep learning architectures, including CNNs and RNNs, applied to this domain, synthesizing key methodologies and datasets. Though early in his career, this review has garnered 2 citations, serving as a foundational resource for researchers exploring end-to-end transliteration systems. His contributions highlight the practical importance of bridging visual and textual modalities, particularly for low-resource languages. Gupta’s work is notable for its systematic approach to evaluating model performance on noisy, real-world signboard images, offering insights into data augmentation and transfer learning strategies. As the field of multilingual scene text understanding grows, Gupta’s research provides a valuable roadmap for developing robust, deployable transliteration tools.