Jade’s Status Report for 4/10/2021

This, week I worked on both the pen tracking algorithm and the hand gesture recognition algorithm.

For the pen tracking algorithm, I tried out a Kalman filter for the acceleration readings from the MPU6050. I tried various parameters but it seemed like the Kalman filter was not doing enough to get rid of heavy inaccuracy in the readings. I met with Bradley and Jenny to discuss alternative solutions, as well as asked our professors for advice. We got a new sensor that incorporates the accelerometer, gyroscope, and magnetometer readings into one. The magnetometer was what we lacked before and we are hoping to make use of this before our interim demo.

For the hand gesture algorithm, I tried several methods to identify the hand but was having a lot of trouble getting it to differentiate between a hand and a face. After seeking advice from our professors and Rashmi, they directed us to Uzair. Uzair told us about mediapipe, a body part recognition library. This has been very helpful, and I’m now able to map hand gestures to user macros, as well as add new ones.