CHECKED OR CARRY-ON? Our inspiration was our love to travel and our distaste for the frustration at the checkpoint at the airport. Many people experience frustration when they are flying. They have to throw away perfectly fine items, such as food or drinks or even harmless valuables because they were not aware of the rules of the TSA, which consisted of over 400 hundred items, all listed in details on their website. No one can up with such a long list. That is why we developed Smooth Pack, an app that utilizes machine learning to detect items that meet or do not meet TSA guidelines.
What it does
Smooth Pack's purpose is to help anyone who wishes to travel by plane that is concerned about what they are allowed to pack. Rather than going through the hundreds of items on the TSA website, this app will help ease your packing and traveling experience. There is also a developing feature to predict the weight of items packed.
How I built it
Smooth Pack is an Android app that utilizes machine learning and does most of its computation on the cloud. We used Google Firebase, a part of Google Cloud Platform, to compute and store data for Smooth Pack. The machine algorithm utilizes Machine Learning Kit / TensorFlow to create accurate classification model for common objects.
Challenges I ran into
The hardest challenge was to learn a new tool to create this app. The members of our team had little experience with Android Studio and Google Cloud services. We had to learn new languages, such as Kotlin and XML in order to create a functional and presentable app.
The time limit was also a challenge. There were many times that we almost gave up because our application kept on crashing, but with the help of amazing mentors we were able to push through.
Accomplishments that I'm proud of
Since this was our team first time working with object detection and Google Firebase, we are proud that we were able to create an app only in 24 hours, considering the fact that three of four of our members are first time hackers. We are also proud of our persistence to complete this application.
What I learned
Everyone on the team learned about Kotlin for the first time and how to utilize it for Smooth Pack. We also learned about Android studio and how to utilize machine learning APIs.
What's next for Smooth Pack
With more products coming out every year, we will be working towards creating custom Machine Learning models that can accurately predict each object for all the travelers out there.