Inspiration
The grandmother of one of our teammates suffers from Alzheimer's, a degenerative disease that results in the breakdown of mental faculties. The purpose of our app was to ameliorate the effect of memory loss on Alzheimer's patients.
What it does
The app uses a machine learning model developed in TensorFlow and the ARKit library to give the user a sleek AR experience. In the scan section, users can tap on objects on their screen, the ML model will identify the object, and the user will have the option to type in the significance of the object. All objects and their significance will be saved in the memories section of the app.
How we built it
We used Swift to create the user interface of the app. The machine learning model was developed in Jupyter, using Python and Tensorflow.
Challenges we ran into
The machine learning model was quite difficult to develop and took the majority of our time.
Accomplishments that we're proud of
We felt quite accomplished after developing a model that actually worked. We were impressed with the final outcome of the app.
What we learned
We learned that even complex and large scale problems such as Alzheimer's can be greatly ameliorated by a simple app such as this.
What's next for VisionScan
Using location to save the position of saved objects to help users locate objects is something we hope to add in the future.
Log in or sign up for Devpost to join the conversation.