Storage Organizer Application

This is the masking of the image with filled in contours
This is the image with boxes and sample measurements per pixel
This is a sample of our front end proposed interface

Inspiration

J - I personally have spent far too long trying to find the best way to organize my library of books. The problem is I have only so many shelves at my place, and not every book fits on every shelf.

What it does

First, it takes a scans of your book spines. Then it uses machine learning to pick out each book and measure its height and depth. Next, each book is compared against each available shelf until it finds the smallest possible shelf that will fit it. Once either all shelves are filled or all books have found their place, it creates an alphabetical list of books for each shelf. These lists can then be used to send your books directly to their spots, plus they can be indexed so you can easily search for a book you're looking for later.

How we built it

The most difficult part was creating a program that can look at an image and can be able to tell you not only where the books are, but also give information about it.

Challenges we ran into

As individual members we were on our way to creating feature complete elements of our program, but issues arose when it came time to make sure if those parts would fit together successfully. Thankfully, we were able to collaborate and work to better understand how each other's code worked so as to more successfully allow them to play off each other. Accomplishments that we're proud of J - I am just proud to have come together over 30 hours and built a feature functional program. An - I am proud about completing my first hackathon and not giving up. Pau - It was my first hackathon and I learned a lot about machine learning and OpenCV. Cash - I am proud of setting up my first IDE in python and completing my first hackathon. I am also happy with how much I learned about front-end development and machine learning algorithms.

What we learned

We have learned valuable lessons on preventing scope creep (planning for creating an application we can comfortably build in our allotted time), practical hands-on experience with common Python features such as Flask, R-CNN, OpenCV, and TensorFlow.

What's next for Storage Organizer Application

The first step would be proper online integration. That was simply not within scope for our skill level during the allotted time. We believe next step would be re-configuring the sorting algorithm to allow for three-dimensional objects, massively scaling up the product's applicability.