Deep Learning for Computer Vision Project

Deep Learning class was one of the best experiences I had at Northeastern University hands down. This is a quick glimpse of one of the projects that we did in class. 

We used Google CoLab and Tensor flow to achieve 82.4% for Cifar 10 dataset and 57.91% for Cifar 100 dataset. My take away in this project is primarily to understand the dataset deeply, followed by the problem statement and playing around with various regression functions, activation function and optimizers which can improve accuracy and predict better in real time. We have Developed an CNN(Convolution Neural Network) application on the other side for fun to predict various animals names, buildings, flowers, fruits etc and all of this can we used to build a recommendation systems  in tracking human actions, quick visual educational learning for children(AKA if we can only plug this in an AR app), build faster building in construction, Interior designing made easy for customer to visualize work and home spaces, health care is bit industry by itself. 

I see this project as a frontier and a start for many focussed based applications that can improve people life's and probably enhance their learning and creative part of their minds. 

Please email me to personally mentioned on the home page to see a detailed report or we can work together or exchange notes!