Repo for our Deep Learning Project 2022 Image Caption Deep Neural Network Model (ICPNNM)
In the following repo we have tried designing our own Image Caption Generator by constructing a Deep Learning Auto-encoder model. Images and captions are gathered from the Flickr8k image dataset from Kaggle "https://www.kaggle.com/datasets/adityajn105/flickr8k" All code are written in Python using Pytorch.
All Py-scripts are stored in the main folder. The 'data' folder consists of the 'images' folder with images and 'texts' with labels and corresponding image-path. Finally the 'features' also exist which holds the feature vector that has been fed through the CNN.
--DeepLearningProject | Data-------------| Images | Texts | Features | Own_Images -----------------| | Pitch | Results
You are welcome to share the work we have done by citing us.