NAC-TCN: Temporal Convolutional Networks with Causal Dilated Neighborhood Attention for Emotion Understanding

Accepted to ICVIP 2023!

Alexander Mehta & William Yang

We propose a method known as Neighborhood Attention with Convolutions TCN (NAC-TCN) which incorporates the benefits of attention and Temporal Convolutional Networks achiving comparable or state-of-the-art performance over TCNs, TCAN, LSTMs, and GRUs while requiring fewer parameters on standard emotion recognition datasets.

Resolving Voltage Peaks in Particle Beam Microscopy

BU RISE

We propose a method to resolve voltage peaks in particle beam microscopy using a combination of estimators and machine learning.

Help the Blind See: Assistance for the Visually Impaired through Augmented Acoustic Simulation

Alexander Mehta & Ritik Jalisatgi

We propose a blind assistive device that uses echolocation based algorithms to allow naviagation. We publish device schematics, code, and a paper detailing our work.