Created on 30th January 2021
•
Our main objective is to help visually impaired people by achieving better visual understanding using image captioning.
Also the project will be useful to tackle problems in field of road safety, defence, healthcare by extracting the action (Caption ) taking place in an image.
Our main challenge was deploying the model on github as the size was too large for github. So, we deployed it on Herokuapp platform, in that also we weren't able to upload our best model for image captioning due to large size so we had to upload the model with minimum size. Also, due to lack of proper GPU cloud, the execution speed is quite less.