A short video that gives an intuitive explanation of what is a convolution operation (without the math): https://www.youtube.com/watch?v=QmcoPYUfbJ8
rsha256
How does this work with images or matrices? I understand the scalar case but with images (presumably which are 3-tensors of int as you have RGB channels) or even a reduced example of convolution with a matrix (which is given on a later slide) I'm kind of confused on how that would work
irina694
@rsha256 I believe convolving an image with a kernel is equivalent to applying a filter (i.e. blurring the image, etc.)
A short video that gives an intuitive explanation of what is a convolution operation (without the math): https://www.youtube.com/watch?v=QmcoPYUfbJ8
How does this work with images or matrices? I understand the scalar case but with images (presumably which are 3-tensors of int as you have RGB channels) or even a reduced example of convolution with a matrix (which is given on a later slide) I'm kind of confused on how that would work
@rsha256 I believe convolving an image with a kernel is equivalent to applying a filter (i.e. blurring the image, etc.)