https://github.com/varadbhogayata/visual-question-answering

The Intelligent system which gives answer to an Image and a Natural Language Question related to image
https://github.com/varadbhogayata/visual-question-answering

cnn-model computer-vision lstm-neural-networks natural-language-processing

Last synced: about 1 year ago
JSON representation

The Intelligent system which gives answer to an Image and a Natural Language Question related to image

• Host: GitHub
• URL: https://github.com/varadbhogayata/visual-question-answering
• Owner: varadbhogayata
• Created: 2018-03-30T10:09:38.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2020-06-08T18:03:40.000Z (about 6 years ago)
• Last Synced: 2025-04-15T03:42:11.573Z (about 1 year ago)
• Topics: cnn-model, computer-vision, lstm-neural-networks, natural-language-processing
• Language: Python
• Size: 59.1 MB
• Stars: 8
• Watchers: 2
• Forks: 5
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Visual-Question-Answering

The Intelligent system which gives answer to an Image and a Natural Language Question related to image. This is a simple updated demo of VQA by [VQA_Demo](https://github.com/iamaaditya/VQA_Demo/) which uses pretrained models (VGG, VQA) to answer a given question.

# Model Architecture

The arhitecture behind the implementation is based on the model proposed in this[paper](https://arxiv.org/pdf/1505.00468v6.pdf)

![LSTM + Q Model](https://github.com/varadbhogayata/Visual-Question-Answering/tree/master/Images/LSTM+Q.png)

# Dependencies

1. Keras

2. Tensorflow

3. Scikit-learn

4. Spacy

5. VGG-16 pretrained weights

# Run

Put image in the data/ directory to test the image on this model, after run this file: 

python VQAdemo.py