JStehouwer / FFD_CVPR2020

111 stars 22 forks source link

attention map and gt mask matching #1

Closed neverUseThisName closed 4 years ago

neverUseThisName commented 4 years ago

Hi, thanks for your work. I have a small question. The attention map size is 19x19, but the gt mask (diff image) is 299x299. Are they matched by downsampling gt mask?

JStehouwer commented 4 years ago

Yes. The gt mask is 299x299, but is converted to greyscale, resized to 19x19, and thresholded (using a threshold value of 0.1 assuming an image is in the range [0,1]) so that it is binary.