Bengali Sign Language (BdSL) Classification
Published:
This is a team project for the CS 4641 Machine Learning course at Georgia Institute of Technology. I trained various Convolutional Neural Networks (CNN) models, including AlexNet, ResNet50, and MobileNetV2, using preprocessed data. Notably, the MobileNetV2-based model achieved an accuracy of 92.8%, surpassing previous literature by Rafi et al. (2019) and being 98% smaller in size.
See https://cbc32.github.io/ for details.
