Description as a Tweet:

Benefitting from modern technologies, our program can interpret the world as we do.

Inspiration:

Our end result was a culmination of multiple previous ideas ranging from AI search to endangered animal protections. All topics shared the use of AI for the common good, so we compromised on a visual AI that could be morphed to aid non-native English speakers and developing children to expand their vocabularies.

What it does:

Our project is a web application that utilizes machine learning to recognize an object in an image and returns the algorithm's predicted result.

How we built it:

We used the Google Cloud Platform to host a virtual machine for our web application to run from. We implemented Google's machine-learning Vision API using Python to construct our web application's back-end. Our front end was developed using HTML/CSS in conjunction with Python Flask to intertwine the separate ends.

Technologies we used:

• HTML/CSS
• Python
• Flask
• AI/Machine Learning

Challenges we ran into:

It was difficult learning new languages such as Python for some members and HTML/CSS for all of us.
Google Cloud took the group a long time to set up properly and we were forced to learn to use the
command prompt to set it up. Implementing the Google Vision API with the python code was pretty hard as well due to the addition of keys that were required for the API to run and track our requests.

Accomplishments we're proud of:

After hours of gaining knowledge from class and YouTube, we got an idea and we pulled through. Having never made a web application before, it was surprisingly difficult to create a simple one. We are proud to announce that kids will be learning how to efficiently upload pictures and use their senses to determine what the picture is alongside our Google algorithm.

What we've learned:

We learned how to use Python with HTML/CSS as well as how to connect it to APIs (Google Vision in this case). In the future, this knowledge and valuable experience can be used to efficiently create a project rather than having to scramble to find what we wanted/had to do.

What's next:

Our next goal for this project is to turn the results from the image into searching products on the web (e.g. Pinterest) for similar ideas and picture examples. Another goal that was discussed was to take a picture (almost like Google Lens) of a person's outfit and return a shopping list of the same clothing items.

Built with:

Python, HTML/CSS, Google Cloud Computing, Google API, Flask, Jupyter, Ubuntu Apache2

Prizes we're going for:

• Best Software Hack
• Best Web Hack
• Best AI/ML Hack
• Best Beginner Software Hack
• Best Mobile Hack
• Best Use of Google Cloud
• Best Domain Name from Domain.com
• Best Space App

Ishan Bhardwaj
Emily Miller
Dong Kang
Evan Rovelli

Table 16