[res] image.convolve([dst,] src, kernel, [mode])
Convolves Tensor kernel over image src. Valid string values for argument mode are :
full : the src image is effectively zero-padded such that the res of the convolution has the same size as src;
When running with mode = 'full' I'm getting a bigger output.
Test script:
require 'image'
local src = torch.Tensor(3, 200, 200)
res = image.convolve(src, image.gaussian(), 'full')
print("src", src:size())
print("res", res:size())
Results:
➜ lua th bug.lua
src 3
200
200
[torch.LongStorage of size 3]
res 3
202
202
[torch.LongStorage of size 3]
From the documentation :
When running with mode = 'full' I'm getting a bigger output.
Test script:
Results: