An example of a very simple 2D convolution is a "blur" convolution function in shader programming. For some indices i,j, we could make a very simple convolution as following:

output[i][j] = (input[i-1][j-1] + input[i-1][j] + input[i-1][j+1] + input[i][j-1] + input[i][j] + input[i][j+1] + input[i+1][j-1] + input[i+1][j] + input[i+1][j+1]) / 9.f;

This is similar to the slide, but for the 2D case. Assuming the output and input arrays were colors, this would, for each pixel, take the average of the color of the pixel and that of every adjacent (including diagonal) pixel.

This comment was marked helpful 0 times.