Khadiza Tul Kubra
Papers
1
Total Citations
2
H-Index
1
About
Dr. Khadiza Tul Kubra is a researcher at the forefront of multilingual scene text detection and recognition, with a particular focus on low-resource and bilingual contexts. Her most cited work, published in 2025, tackles the challenging problem of detecting and recognizing bilingual Urdu and English text in natural scene images. By combining a Convolutional Neural Network–Recurrent Neural Network (CNN-RNN) architecture with a Connectionist Temporal Classification (CTC) decoder, she developed a robust system capable of handling the complex, cursive script of Urdu alongside English in real-world settings like signboards and navigation boards. This contribution is pivotal for modern applications such as language translation for foreign visitors, robot navigation, and autonomous systems. With 2 citations already, her work is gaining traction in the computer vision and document analysis communities. Dr. Kubra’s research addresses a critical gap in multilingual text understanding, demonstrating how deep learning can bridge linguistic diversity in visual environments. Her innovative approach underscores the importance of inclusive AI systems that serve multilingual populations, making her a rising voice in scene text recognition and its practical deployment.
Research Focus
Key Achievements
Top Papers
- 1