Padding
● Padding is a simple trick to add extra pixels to the edges or boundaries of the
image so that the effective size of the image gets increased and there is no
loss in the original number of pixels from the image. The boundary edges are
preserved during convolution.
● One of the basic types of padding is called Zero Padding. In this, we add zeros
to all the edges of an image to prevent the loss of the original number of pixel
values of the image.
● We usually perform zero padding on the corners of the tensor. We do this to
get a size of output that is the same as that of the input.
● Here in the below image, we observe that we have padded both sides of the
tensor with Zeros.
faizanbaig@[Link]
TYQJ7EB1OR
This file is meant for personal use by faizanbaig@[Link] only.
Sharing or publishing the contents in part or full is liable for legal action.