search

marco-rudolph / cs-flow

This is the official repository to the WACV 2022 paper "Fully Convolutional Cross-Scale-Flows for Image-based Defect Detection" by Marco Rudolph, Tom Wehrbein, Bodo Rosenhahn and Bastian Wandt.
111 stars 30 forks source link

Why only the first 304 channels are used #16

Closed ZwkAlex closed 2 years ago

ZwkAlex commented 2 years ago

the efficentnet featrure extractor output channels is 512, but inputNode only used the first 304 channels.

marco-rudolph commented 2 years ago

This is because we do not use the last convolutional layer, but an intermediate one (index 36). This layer has 304 output channels.

ZwkAlex commented 2 years ago

Thanks for the explanation