Skip to content

The repo contains a machine learning model to generate captions for an image. The model is deployed on web using streamlit.

Notifications You must be signed in to change notification settings

ABHISHEKgauti25/ImageCaptionGenerator

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 

Repository files navigation

ImageCaptionGenerator

The repo contains a machine-learning model to generate captions for an image. The model is deployed on the web using Streamlit and ngrok. The code of the model can be found in model_code.ipynb while the deployment code code is in deploy_code.ipynb.

Dataset

The model is trained on the flickr8k dataset. The dataset consists of around 8000 images with an average of 5 captions per image.

Model Architecture

I am using a Pre-trained VGG16 model to compute features present in an image. Skipping the last softmax activation layer of VGG16, the previous layer's output is directly fed into the model below.
model

Note: I have uploaded a trained version of the above model architecture which can be found here:https://github.com/ABHISHEKgauti25/Webapp-imgCap/tree/main/trained%20models Trained for nearly 50 epochs, this was able to achieve a Bleu score of 0.51.

About

The repo contains a machine learning model to generate captions for an image. The model is deployed on web using streamlit.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published