-
DUCK
-
DUCK
-
GO
Inspiration
We were inspired by the theme of the hackathon, and wanted to create awareness and spread information about the ducks and their various species.
What it does
The app allows user to take an pictures of duck to identify the species of the duck, providing information about the duck. The app also lets users to take a snapchat-like filter from the image to create an interactive user learning experience.
How we built it
We utilize Computer vision AI to be able to tell apart different kinds of ducks, based on the Flash framework to provide an interactive UI experience.
Challenges we ran into
We dodged various challenges while building the app, such as training the AI model to differentiate ducks took a lot of time and we have to spent time fine tuning the model as well for accurate predictions. We also wanted to build an app for Android and IOS using Flutter but due to time constraints we weren't able to finish.
Accomplishments that we're proud of
We are proud of our product, and the step we took to build it. Design Computational vision AI s, defining a framework to host the model and utilizing interfaces such as cameras and creating filters.
What we learned
We learnt more about Computational Vision AI/ML libraries, using OpenCV. We learnt about using Flutter to build an app even though we weren't able to finish.
What's next for DIY Duck Face
We would like to go back and build our apps to various platforms, also increasing range of animals and birds detectable with the app, create more user interactive filters and AR features for a interactive user experience.
Log in or sign up for Devpost to join the conversation.