search

MarvinTeichmann / tensorflow-fcn

An Implementation of Fully Convolutional Networks in Tensorflow.
MIT License
1.1k stars 433 forks source link

Problems with gray images #24

Closed jiao0805 closed 7 years ago

jiao0805 commented 7 years ago

Hi Mr.Marvin Teichmann I found this in the usage that: The images is a tensor with shape [None, h, w, 3]. Where h and w can have arbitrary size.

My training data is gray images with no rgb channels. Can I still use this code and vgg to train them?

MarvinTeichmann commented 7 years ago

Yes, you can. Make sure that you load or convert the image as 3 channel tensor, e.g. by calling scipy.misc.imread(image_file, mode='RGB'). The input producer of KittiSeg works out of the box with this code for greyscale images.

jiao0805 commented 7 years ago

Thanks a lot for your help！ ^-^