GlassSlipper.ai

Inspiration

Getting ready is hard, but it doesn't have to be. We wanted to look and feel our best so we created a tool to do so.

What it does

Simply upload a picture of your favorite celebrity and you will receive makeup, hair, and jewelry tips and tricks to emulate your fav celebrity's looks.

How we built it

We built it using HTML, CSS, and JavaScript for the frontend. We used FastAPI, Amazon Rekognition, and OpenAI to analyze the celebrity's face and features, and feed it into the OpenAI API to output a response.

Challenges we ran into

We originally wanted to analyze the undertones, hues, etc. of the user's face but the APIs that could do this were all quite expensive. So we opted to use the free tier of Amazon Rekognition. Connecting all the elements and parts was difficult to make it all work.

What's next for GlassSlipper.ai

1. Address issues of API calls: Have each user create an account to track API calls – can limit a certain number per month. The number of calls per user will be stored in a database so that the user will notified of when they reach the limit.
2. Upload a picture of the user’s face so that the tool can tailor suggestions the user’s skin tone and facial structure
3. Work on the prompt engineering into the Open AI API so that GlassSlipper outputs tailored and more detailed responses

Built With

• amazonrekogitionapi
• css
• fastapi
• github
• html
• javascript
• openai
• openaiapi
• python
• vscode