Low Cost Static Gesture Recognition System Using MEMS Acceleromters
* Designed and constructed a glove that recognized static gestures.
* Developed a light weight algorithm to identify alphabets in American Sign Language.
* Presented in Global IoT Summit, Geneva, Switzerland.
Design Optimization of Activity Recognition System on An Embedded Platform
* Modeled an activity recognition system using just accelerometer data to identify activities.
* Engineered features to be minimalistic yet effective and worked on 3 datasets.
* Deployed the above system on Raspberry Pi Zero.
* Presented at Future of Information and Communication Conference, Singapore in April 2018.