search

HasnainRaz / FC-DenseNet-TensorFlow

Fully Convolutional DenseNet (A.K.A 100 layer tiramisu) for semantic segmentation of images implemented in TensorFlow.
MIT License
123 stars 41 forks source link

the outputs are a vertical line #18

Closed yangyang9706 closed 5 years ago

yangyang9706 commented 5 years ago

I comment out the normalize function，but the outputs are just a vertical line and blank.

HasnainRaz commented 5 years ago

In the dataloader, under the resize image and mask function, use the "tf.image.resize_nearest_neighbor()" function for the mask. I assume its an issue with the bilinear interpolation in the resizing. Let me know if it works.