Skip to content

Remove, Replace and Fill anything in an image with Stable diffusion inpainting. Generate social media captions and Hashtags for the edited image with LLMs

Notifications You must be signed in to change notification settings

shreyassks/Content-Genie-AI-Image-Editor

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Social Media Content Creator - Content Genie

An AI based Image Editing and multilingual social media caption and hashtag generator

Content Genie Demo Video

content_genie.mp4

Google Slides Presentation

https://docs.google.com/presentation/d/1pbDHA7DIuM8BArwQZqm8eD-PZJRn28eg/edit?usp=drive_link&ouid=113737295631185544437&rtpof=true&sd=true

Product Features in Content Genie

capabilities.png

Model Interpretability using Diffusion Attentive Attribution Maps (DAAM)

heatmap.png

Description

In today's competitive business landscape, engaging customers through effective marketing content, especially on digital channels, is crucial for driving business goals and building brand loyalty. However, many customers often struggle to create compelling marketing materials that resonate with their target audience. They may lack the resources, skills, and time to produce high-quality content consistently. It is frequently observed that Social Media Managers usually work with agencies that create the content for them. Any tweak in the content requires to and fro between the marketing managers and the agencies, which is not cost effective both in terms of money and time. There could be time-sensitive campaigns that would require the Social Media Managers to modify the content immediately.

The challenge for this hackathon is to develop a platform or solution that empowers customers with real-time, AI-powered image editing capabilities allowing them to customize and enhance the generated images on-the-fly, with options to remove, replace or refill specific objects using intuitive text prompts. The platform would also employ a style-based caption generation model, automatically producing contextually relevant and stylized captions that resonate with the visuals. The solution would be user-friendly, scalable, and accessible to businesses, regardless of their technical expertise.

Features

Edits images and recommends suitable social media captions and hashtags

  1. Removes any object in the image
  2. Fills the masked object corresponding to the text prompt
  3. Replaces the background corresponding to the text prompt
  4. Recommends Captions and Hashtags suitable for the selected image

Installation

To install this service there are 2 ways. Often, running the service locally allows you to easily change and test the code. If you're ready for productizing, run the service as containerize service.

Run service locally

Install virtual environment

Step 1 :- Creates a Virtual Environment named "venv"

python3.10 -m venv venv

Step 2 :- Activate the Virtual environment

source venv/bin/activate

Step 3 :- Install required packages

pip install -r requirements.txt

Step 4 :- Enable sso - Go to home directory and run below command

aws sso login --profile sso_profile

Step 5 :- Allow SSO on browser

Select Open
Check that the code is auto filled then Click Next 
Click Allow

Step 6:- Run the below command from app folder

uvicorn main:app --host 0.0.0.0 --reload

Step 7:- Now you can use the integrated docs at:

http://127.0.0.1:8000/docs

![Alt text](assets/fastapi landing page.png)

Directory Structure of Backend Services

About

Remove, Replace and Fill anything in an image with Stable diffusion inpainting. Generate social media captions and Hashtags for the edited image with LLMs

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages