I read that image pyramids are used for deep dreaming. The image is feed through the network multiple times and is rescaled during the process, to capture features in different scales. But how can the neural network handle the different input sizes? Are the inputs padded with zeros to match the network input size?
Asked
Active
Viewed 27 times
1
-
Can you show where you saw this? – user253751 Mar 28 '23 at 12:57
-
@user253751 Yes, sure. https://m.youtube.com/watch?v=6rVrh5gnpwk at timestamp 6:31. – lev1248 Mar 29 '23 at 10:48
-
I see. Perhaps they were using the convolutional part of the network, which is size-independent – user253751 Apr 01 '23 at 09:16