Impliments Residual Neural Networks to improve performace in addition to reducing computational time significantly. This model outputs values between 0 and 5 based on the SIGN represented in the image. Y= 0 if hand shows 0 Y=1 if hand shows 1 Y=2 if hand shows 2 continued until 5. This model has a test accuracy of 87% and can be further improved by implimneting regularization due to high variance. The performance can be further improved by using Principal Component Analysis(PCA). Batch Normalization should be implimented on Res-Nets to improve performance further. Libraries used include: keras,scipy, matplotlib and tensorflow