What it does

Our program compares two different hand movements (one after another) and displays the percent similarity in motion. The purpose of this program is to teach hand movements in critical fields that rely on precision motor skills. For example, it might be implemented by medical students for medical simulation to teach the ins and outs of surgery and suturing in a safe environment. It may also be used in hospitals and the food-service industry to determine proper hand-washing protocol.

How we built it

Using Leap Motion, a hardware tool with a built in IR camera and their developer API library to record and compare the properties of two different hand movements in Java using mathematical models.

Challenges we ran into

We encountered issues with hardware limits and getting precise hand vector information to create a good comparison.

Accomplishments that we're proud of

Our program works with relatively high accuracy and precision.

What's next for Real-Time Gesture Comparison

Our hope is to implement a friendlier GUI. We would like to create a library of saved gestures in order to compare their hand movements to a library to eliminate the need of a instructor present. We would also like users to create custom gestures that can be created via open source classification training models in Python/ML.

Built With

Share this project: