Image Captioning using Deep Learning
The code has been written using the Keras deep-learning library with Tensorflow backend.
Following libraries are required to run the code:
• os • requests • zipfile • io • numpy • datetime • matplotlib • pickle • nltk • re • BeautifulSoup (bs4) • Keras • Tensorflow
The above libraries either come pre-installed or can be installed using the pip command in a Python virtual environment (virtualenv).
The Flickr8K dataset can be downloaded from the following links:
• Images: https://github.com/jbrownlee/Datasets/releases/download/Flickr8k/Flickr8k_Dataset.zip • Text: https://github.com/jbrownlee/Datasets/releases/download/Flickr8k/Flickr8k_text.zip
According to the location of the files, the folder and data paths mentioned in the code need to bemodified.
The whole code is accessible on Google Colaboratory via the following link: https://colab.research.google.com/drive/1ZOA195yLGo2OtVDMALI_pc5mq2lSeMVg?usp=sharing