search

shekkizh / FCN.tensorflow

Tensorflow implementation of Fully Convolutional Networks for Semantic Segmentation (http://fcn.berkeleyvision.org)
MIT License
1.25k stars 527 forks source link

relu5_3 #46

Closed yossitsadok closed 7 years ago

yossitsadok commented 7 years ago

Hello,

Is there a particular reason why the Inference continue from the "conv5_3" level of the vgg_net and not from the "relu5_3" ? I mean teh input to "POOL 5" is "conv5_3" and not "relu5_3"

What am i missing?

Thanks!

shekkizh commented 7 years ago

My take is that relu layers by themselves aren't going to affect the inference in this case since they are merely thresholding functions. Not sure if that answers your question but am closing this issue for now.

varungupta31 commented 2 years ago

@yossitsadok https://github.com/shekkizh/FCN.tensorflow/issues/94#issuecomment-1108473347