Skip to content

Latest commit

 

History

History
24 lines (17 loc) · 1.88 KB

README.md

File metadata and controls

24 lines (17 loc) · 1.88 KB

Image Caption Generator

After researching multiple models I came across BLIP by salesforce and tried implementing the model on a Huggingface space. I received successful results but the gneerated caption was simply a description of the image.

https://huggingface.co/docs/transformers/model_doc/blip https://huggingface.co/spaces/rimssss/bliptrial image

The issue was that the caption was only the description of the image and it wasn't catchy or captivating. So, to the solve the above problem I further passed the description to a GPT model along with an option for the user to submit a prompt and accordingly generate the desired caption.

Image2 image Image3 image Image1 image

And thus the model was able to generate "catchy", "innovative" and " exciting" (as prompted by the user) captions which can easily be used as an actual caption for social media posts.

Requirements:

-Transformer

-Torch

- Flask