Visual Question Answering Using Deep Learning

Ask any question about the given image and let the model answer it 

Project Description 

This project focuses on building a Visual Question Answering (VQA) web application that allows users to upload an image and ask natural language questions about it. The system leverages a pre-trained Vision-and-Language Transformer (ViLT) model to jointly understand both visual and textual inputs and generate accurate answers. The application is deployed using a Flask backend, with a responsive frontend that enables seamless interaction through a web interface. 

Features 

• Upload an image and ask questions in natural language

• ViLT-based Vision-Language Transformer model for multimodal understanding

• Real-time answer generation from visual and textual input

• Web-based interface for easy user interaction

• Backend integration via Flask REST APIs

• Suitable for AI, deep learning, and NLP-based academic projects

Tech Stack

• Python

• Vision-and-Language Transformer (ViLT)
  
  

• Hugging Face Transformers

• Flask (Backend Framework)

• HTML, CSS, JavaScript (Frontend)

• Other supporting Python libraries

System Requirements

• Python 3.10+

• Anaconda

• No GPU is required

Deliverables 

• Project Code

• Setup over video call 

• WhatsApp support