Inspiration
We were inspired by our grandparents' mutual love for plants. We realized we could integrate this into our fitness app as a way to entice and maintain user participation.
What it does
This is a friendly and accessible user interface for senior citizens to log their workout practices in a fun and entertaining manner. Here their “plants” thrive, along with their physical health.
We used artificial intelligence to create a workout plan for senior citizens based on what they want to focus on, and how much time they have available. We integrate this program into the Planting Positivity app, as an activity option for users who want personalized workouts.
How we built it
We built the motivational plants interface on Scratch and used various variables and sprites to generate the output we wanted. For our workout generator we used python on vscode, and used OpenAI's library to get access to the gpt-3.5-turbo and dall-e-3 models.
Challenges we ran into
Despite our team being proficient in Java and JavaScript, we learned that we couldn’t make an effective user interface using Java with our skillset.
The AI model used for the image generation in part 2 (dall-e-3) does not always contain fully realistic images and contains spelling errors.
The AI model that creates the workout from the inputs (gpt-3.5-turbo) can contain workouts not entirely effective for someone, or that could be too hard for senior citizens.
Accomplishments that we're proud of
We all had very busy schedules this weekend but still wanted to participate in the hackathon despite the large time commitment. We are proud that we were able to work around our numerous prior commitments and still come together as a team. We also were able to use our individual skillsets to create a more effective overall product.
Integrating AI with our code was newer to us, and we are glad that we were able to effectively create image outputs.
What we learned
We learned how to create outputs using artificial intelligence models. We also had never used Scratch before and learned the new programming platform while building our app.
What's next for Planting Positivity
We can try to use JavaScript, HTML, and CSS to create Planting Positivity in a webpage format.
Add more accessibility features for senior citizens who may struggle with hearing/sight impairments.
Find a better AI image generation model than dall-e-3.
Fine tune the GPT-3.5-turbo model with training data to receive more consistent and effective results.
Built With
- python
- scratch
Log in or sign up for Devpost to join the conversation.