Draw and Scan

• 

  Trial run with our webpage sketch

• 

  another run, with clickable menu buttons

Inspiration

We both suck at UI design, but sometimes we have an idea that comes to mind that we think would be good.

What it does

You draw a picture of what you envisioned on a whiteboard or a sheet of paper, and upload an image of it to generate the HTML code and an image of what it would look like.

How we built it

We used Google Gemini and JavaScript

Challenges we ran into

None of us has ever used JavaScript before, so we used prior coding knowledge (C and Python) as our baseline. The images that Gemini produced would include explanations of the image generation in the image, which led to improper formatting. We had to isolate the text generation and remove it from the image.

Accomplishments that we're proud of

It works :) (to an extent)

What we learned

We gained experience in implementing Gemini API into our project, which can be applied to furthur project relating to LLM. We also learned how to use GitHub to publish our project.

What's next for Draw and Scan

It was a fun little project. We'll see :)

Built With

• gemini
• github
• javascript
• node.js