We are extremely grateful to have been chosen as one of the 6 finalists. Thank you for hosting a memorable first hackathon QHacks!
Our team wanted to create something that would be useful in today's world while also aiming for the Best Use of Google Cloud Award. Inspired by the theme, "The New Wave of Tech", we wanted to push boundaries of what a simple mobile app could do . . . and thus we made Lucid, an app with the ability to help the blind.
What it does
As explained, Lucid aims to make life easier for the blind. It's main function is to use the phone's camera and text-to-speech to tell the user what objects are in front of them (object recognition) or what the text in front of them says (text-recognition). Ex: Grocery shopping, blind people can't tell if an apple is red or green! We also have incredible UI features built in like physical/tactile feedback on button clicks, text-to-speech to describe the activity that appears on the screen, voice command startup, speech speed customization, and many other features so that Lucid is perfectly curated for users without sight.
How we built it
Backend: Using Google Cloud's Vision API, Machine Learning Kit (machine learning for mobile developers), Java, and Android Studio
Frontend: XML and Android Studio
Challenges we ran into
Going into the Hackathon, we lost 2 members. So that left only me to build the app and my partner, Andy Ho, to develop the pitch/demo materials. In the end it all worked out and the thrill of the challenges really pushed us.
Accomplishments that we're proud of
Creating and learning so much in such a short amount of time, especially after the challenges we both ran into.
What we learned
We learned about Google Cloud and their resources. We also found it very interesting to learn how to make use of cameras and text-to-speech in Android Studio.
What's next for Lucid
There are many things Lucid can improve on like it's accuracy on image recognition, more language support, make use of Google Cloud's translation API to convert text from one language to speech in another, and many more!
Team Channel #2 (coldcabbage#8622 and greenbean537#6339)