3 ms·
Thank you for your comments! For sure, the CNN is expressive for learning the characteristics of images. However, in this development, I tried to not use deep-l
by C-Naoki 2y ago
Thank you for your comments! For sure, the CNN is expressive for learning the characteristics of images. However, in this development, I tried to not use deep-learning because I believe that it is important to provide fast, consistent results without the need for training data. If you are particularly interested in this app, I would be glad if you could create a pull request to extend the algorithm.
- jdhwosnhw 2y agoThe parent comment said nothing about using deep learning. Convolution is not the same as using a CNN. I interpreted their comment as meaning they used a 2D convolution (presumably a 2D cross correlation, actually) to find regions of overlap
- scottdupoy 2y agoYes you're right it was a 2D cross-correlation which is very analogous to a convolution
- r_hanz 2y agoIf memory serves… the only difference is that one of the kernels being convolved is reversed for convolution.