Inspiration
At the current moment, there does not exist a solution to measure an athlete's agility and reflexivity. By Measuring these statistics an athlete can improve his skill levels to better perform in hir respective field of activity.
What it does
The software tracks and monitors the athletes' movements to calculate different types of movements such as lateral, back and forth, reaction time.
How we built it
We used the following tech for this:
- OpenCV
- Python
- JS
- Postgres
- FLask
Challenges we ran into
we faced lack of hardware support for the tracking of movements. Hence, we had to use our cell phones for signalling the next point. The webcam is not high definition, hence the margin of error. All of the programs run on a single computer right now and hence we faced a lot of performance issues, which can be improved using dedication cpu.
Accomplishments that we're proud of
We are proud to develop this software as none of us had experience with computer vision in such extensive manner. We went ahead with just an idea and transformed it successfully into a proof-of-concept.
What we learned
Agility training can be accessible to people at home. With simple setup and minimal hardware, anyone can track their agility and endurance anywhere.
What's next for AgilityCoach
We can convert this into a full fledged product using the following roadmap:
- Upgrading existing opencv logic to a better image recognition model that tracks the foot and hand instead of using the masking logic
- Changing the simple polling logic of the next point to a low latency one to improve experience. IOT can be used here, creating dedication hardware to be used while training.
- Taking body type, weight, height, age, gender into account while comparing the data.
Log in or sign up for Devpost to join the conversation.