DanceAI

Inspiration

Some members of our group are experienced dancers, while others are interested in learning. Since dancing isn’t very beginner-friendly, we wanted to make learning and improving at dance more accessible. Moreover, people with more experience in dancing can use DanceAI to fine-tune their technique during their own practice sessions.

What it does

Our DanceAI lets users upload a reference dance video and record their own performance for side-by-side comparison. It analyzes limb angles and slopes to deliver feedback and advice for improvement.

How we built it

We split into front-end and back-end teams, each working in its own Git repository. While developing independently, we kept each other informed of our progress and built components that were later merged into a unified whole.

Since we can only submit one Git repository, you can find the front-end Git repository here: https://github.com/JJW68/dance_analysis_frontend

Challenges we ran into

Combining our sections proved more challenging than expected. Throughout the project, we often ran into misunderstandings over specifications, which pressured us to strengthen our communication and collaboration.

Accomplishments that we're proud of

We successfully pitched our project within the time limit, and created a tool we would use and expand upon in the future.

What we learned

We learned that setting clear expectations is crucial, and that allocating sufficient time for merging code is essential, because merging code can be more challenging than writing code.

What's next for DanceAI

Our next step is to launch an MVP to collect data. We’ll then use this data to train and improve our machine-learning model, enhancing the quality and personalization of our feedback.

Built With

• python
• typescript