Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{192832,
        author = {Puli Maheswari and Kodiganti Likhitha and Bariki Saroja and E M Lakshmi and Dr C V Madhusudhan Reddy and M sivamma},
        title = {AI-Based Picture Translation App},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {9},
        pages = {2214-2220},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=192832},
        abstract = {In an increasingly globalized world, the ability to extract and translate text from images such as posters, signs, and documents is crucial for accessibility, information dissemination, and cross-cultural communication. This paper presents an innovative multilingual text and image extraction system designed specifically for poster content using advanced artificial intelligence techniques. Our system integrates Qwen2-VL- 2B-Instruct, a state-of-the-art vision- language model, with Google Translator to provide seamless text extraction and translation across 15+ languages including English, Hindi, Spanish, French, German, Chinese, Japanese, Korean, Arabic, Russian, and several Indian regional languages. The system is built using Streamlit framework, providing an intuitive web-based interface that enables users to upload images, extract text, perform real-time translation, and search within extracted content. Experimental evaluation demonstrates that our system achieves high accuracy in text extraction (97.3% F1- score) and maintains translation quality comparable to human translators while processing images in under 3 seconds on average. The system's ability to handle mixed- language content, particularly Hindi-English bilingual text, makes it particularly valuable for multilingual regions such as India. This work contributes to the field of AI-powered document processing by demonstrating an effective, scalable solution for multilingual OCR and translation with practical applications in education, tourism, accessibility, and information access.},
        keywords = {Multilingual OCR, Vision-Language Models, Text Extraction, Machine Translation, Qwen2-VL, Poster Analysis, Cross-lingual Processing, Artificial Intelligence.},
        month = {February},
        }