Test: Clip vs. Non-clip about alphagan-matting HOT 1 OPEN

I just put the whole image into the generator for the non-clip case, and it could work in some cases. Except the clip or non-clip method, I also tried to padding the input to (n * 320, m * 320) with constant value like 0 for trimap. But the result was not so good as the original paper's. Maybe I didn't train the model well. My classmate suggested me to use sliding window on the input image, where the window size is 320*320 and the stride of it could be 200 * 200. Maybe you could try that method.

from alphagan-matting.