• 

  App Logo

• 

  Image used for analysis

Inspiration

GlowAI is an innovative project that combines artificial intelligence with image processing to analyze your face for skin conditions such as acne, and makes recommendations for products based on the severity of the condition.

Note: Due to constraints on sourcing quality data and the cost of hosting a model in the cloud, GlowAI is currently a locally run application. It is trained on the ACNE04 dataset and is limited to identifying acne at this time.

We realized after recording that the screen recording only captured the Chrome window, so we uploaded the image used for analysis in the image gallery of the project.

Features

• Comprehensive product database
• AI model trained on various skin conditions
• Intelligent product and skincare ingredient recommendations

License

This project is licensed under the MIT License.

Contact

For any questions or feedback, please open an issue on this repository or contact the project maintainer at atp2323@utexas.edu.

Acknowledgements

This project was built for the Nosu AI/ML Hackathon. https://nosu-ai-hackathon.devpost.com/

What's next for GlowAI

• Ideally, a larger, higher quality dataset of images containing other skin conditions would allow the model to excel in identifying all types of faces and make more fine-tuned product recommendations. Currently, the model is only trained on the ACNE04 dataset.
• Model hosting! We didn't want to pay for model hosting so everything was run locally, but ideally this would be deployed as a mobile app or a low-friction web-app, allowing for ease of usability to prospective users.
• Frontend refinement; again, this would be best deployed as a mobile app, but due to team issues and time constraints, a basic application was coded up to have something for submission.

Built With

• codebuff
• flask
• html5
• kaggle
• nebius
• openai
• python
• pytorch
• supabase