view article
Figure 3
Schematic of a three-layer MSDNet with cin and cout the number of input and output channels. Blue, green and red solid lines represent 3 × 3 dilated convolutions between all possible pairs of input and intermediate layers, with different dilations assigned to each color. The black dashed lines at the bottom connecting all input and intermediate layers to the output layer represent 1 × 1 convolutional operators, amounting to a linear sum between individual pixels at each position among all non-output layers.

Journal logoJOURNAL OF
APPLIED
CRYSTALLOGRAPHY
ISSN: 1600-5767
Volume 57| Part 2| April 2024| Pages 392-402
https://doi.org/10.1107/S1600576724001390
Follow J. Appl. Cryst.
Sign up for e-alerts
E-alerts
Follow J. Appl. Cryst. on Twitter
Twitter
Follow us on facebook
Facebook
Sign up for RSS feeds
RSS