Skip to content
A NN that generates captions word by word using an ensemble model composed of a VGG16, and 2 trained LSTMs.
Jupyter Notebook
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.gitignore
README.md
helping-the-blind-see.pdf
image_captioning_model.ipynb

README.md

Helping The Blind See

This project is aimed to deliver a prototype of an assistive technology tool using image captioning to help the blind understand the visual environment around them. This is done using an encoder-decoder neural network that builds captions word by word from a given image and associated caption.

Take a peek into the notebook within the repo. It contains alllll the details for creating, training and testing the model.

You can’t perform that action at this time.