Nikronic CoarseNet issues - Githubissues

Nikronic / CoarseNet

Modified version of U-net based on "Convolutional Networks for Biomedical Image Segmentation (Ronneberger et al., 2015)" paper.

https://arxiv.org/abs/1505.04597

MIT License

25 stars 2 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Use feature visualization in the report

#13 Nikronic opened 5 years ago
0
VGG part of loss function cannot be decreased at all.

#12 Nikronic opened 5 years ago
1
Mean and STD values regarding our dataset is required

#11 Nikronic opened 5 years ago
3
CoarseNet is not working as a low-pass filter - Halftone patterns are visible

#10 Nikronic opened 5 years ago
1
RandomNoise

#9 hamed-faraji closed 5 years ago
1
VGG gradients explode significantly after epoch one

#8 Nikronic closed 5 years ago
2
Loss function has been implemented wrongly

#7 Nikronic closed 5 years ago
3
Normalization is required

#6 Nikronic closed 5 years ago
2
Noisy output

#5 Nikronic closed 5 years ago
9
Number of encoder and decoder layers are not match. How to apply skip connections?

#4 Nikronic closed 5 years ago
0
Based on modification mentioned in table 1, x and y dimension on final CL layer, goes 0x0.

#3 Nikronic closed 5 years ago
2
Add ability to detect GPU if exists and use proper device

#2 Nikronic closed 5 years ago
0
Add test modules in separate folder

#1 Nikronic closed 5 years ago
0