English Visual Question Answering: Building a Culturally Relevant Dataset from Image Captions
International Journal of Scientific Research in Engineering and Management · 15.03.2026 · 10 (03) · 1-9
DOI: 10.55041/ijsrem57750 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Scientific Research in Engineering and Management, online erschienen am 15.03.2026. Verfasst von Dr. Md Sirajul Huque und 3 weiteren Autor:innen.
Abstract (Originalsprache)
- Visual Question Answering (VQA) is a challenging multimodal task requiring joint understanding of visual content and natural language questions. Traditional VQA systems rely on complex attention-based architectures demanding significant computational resources and GPU training. This paper proposes an efficient and scalable VQA system using a pretrained CLIP (Contrastive Language–Image Pretraining) ViT-B/32 model for open-ended English language queries. The proposed approach extracts semantically aligned image and question embeddings using a frozen CLIP backbone and combines them through a lightweight Multi-Layer Perceptron (MLP) classifier for answer prediction. Experiments on the VizWiz dataset — a real-world benchmark of images captured by visually impaired users — demonstrate competitive performance, achieving a Top-1 accuracy of 40.6% and Top-5 accuracy of 71.7%, trained entirely on CPU without end-to-end fine-tuning. A Flask-based web application supporting user authentication, image upload, and real-time Top-5 predictions with confidence scores is also demonstrated. Key Words: Visual Question Answering, CLIP, VizWiz Dataset, MLP Classifier, Multimodal Learning, Deep Learning
Zitieren
APA
Huque, D. M. S., kumar, B. D., Mandadi, B., & kumar, C. A. (2026). English Visual Question Answering: Building a Culturally Relevant Dataset from Image Captions. International Journal of Scientific Research in Engineering and Management, 10(03), 1-9. https://doi.org/10.55041/ijsrem57750
Vancouver
Huque DMS, kumar BD, Mandadi B, kumar CA. English Visual Question Answering: Building a Culturally Relevant Dataset from Image Captions. International Journal of Scientific Research in Engineering and Management. 2026;10(03):1-9. doi:10.55041/ijsrem57750
BibTeX
@article{Huque2026,
author = {Huque, Dr. Md Sirajul and kumar, B. Dinesh and Mandadi, Bhoomika and kumar, C. Anil},
title = {English Visual Question Answering: Building a Culturally Relevant Dataset from Image Captions},
journal = {International Journal of Scientific Research in Engineering and Management},
year = {2026},
volume = {10},
number = {03},
pages = {1--9},
doi = {10.55041/ijsrem57750}
}
EUVIMED