Meetings

Found and fixed an issue in the integrated classifier that was causing some of the extra time the added to the latency for detecting images!!  Sadly this was after the demo.. but now we surpass all our requirements.  Time to detect gesture has gone down …

Demo-ed current status on manual mode to Professor Sullivan and Jens Integrated animation with the gesture classifier Spent time yesterday and today morning trying to make classifier work on real images Attempted to train other_classifier.py on an augmented dataset (kaggle + real images) – did …

Decided to use Python pygame instead of Simulink/Matlab for the animation (why we didn’t think of this before is a mystery) – much easier and simple to integrate as well Testing gesture classifier on real images Goal is to have the pipeline ready by Wednesday …

Met with Professor Sullivan and Jens and went over our SOW and Design Review Report feedback. Professor Sullivan suggested a couple very useful things for the mics: Use egg cups + Styrofoam padding + plastic cup padding to make the baffles Try playing with the …

Ordered 4 Adafruit mics from Amazon, emailed Jens and Quinn requesting that they place order asap Most online vendors have stopped shipping Because of this we can only get 4 mics – change plan to hardcode one COMOVO as the one that can do automatic …

Researched directional microphones.  Very hard to find a unidirectional cardioid microphone that fits our needs perfectly.  Most of the ones found are too large and expensive since they’re made for singing/recording purposes. Emailed Professor Sullivan asking for input on whether we should buy one of …

Had a Zoom check-in meeting with Professor Sullivan and Jens to discuss what changes we should make for our project and what our modified requirements should be They can ship us all our parts (We will be emailing Quinn soon.) Need to order directional mics …

First Zoom meeting today! Met with Jens and clarified some logistical things. Talked a little about how we might refactor our project design and which parts can and should be done remotely. Emailed Professor Sullivan our availability. Plan to work on the project as usual …

picked up the cameras and motors! ordered a usb to usb-c cable for Neeti installed OpenCV version running Python 3 on RPi Sam instead of using virtual environments Figuring out how to process keystroke input on RPi without using GUI TO-DO: waiting for the motor …

Worked for many many many hours yesterday and today to finish writing and format the design review report. Discussed plans for this week (before spring break) – we want to begin work on the ML model for gesture detection and begin manufacturing the data sets …