Digit Image Denoiser

• Task : De-noise Augmented Image (Noised, Rotated, Transformed)
• Trained and Implemented De-noising Auto-encoder and De-noising Variational Auto-encoder with encoder and decoder in Res-Net Style from scratch using PyTorch.
• Trained a condition Variational Auto-encoder to generate one of the classes of MNIST Dataset at Inference time given class label (Generating De-noised Image).

Digit: 0

Digit: 1

Digit: 2

Digit: 3

Digit: 4

Digit: 5

Digit: 6

Digit: 7

Digit: 8

Digit: 9