Why do you use an image cropping function instead of resizing?

GeorgeSeif / Semantic-Segmentation-Suite

Semantic Segmentation Suite in TensorFlow. Implement, train, and test new Semantic Segmentation models easily!

2.51k stars 878 forks source link

Why do you use an image cropping function instead of resizing? #248

Closed ghost closed 3 years ago

ghost commented 3 years ago

I don't understand how you got the output images that you have included in the images file as the only outputs I am getting when training and testing are cropped versions of the original at 512 x 512, how can I change this so that the outputs are of the full image and are of original resolution?