jingzhang617 / scribble_saliency Goto Github PK

why is it you never get a code right first time and make a easy to run you code so mean errors l use this code MiDaS-2 work fine first time no problem and easy to run and made a readme file which gives you where to put your image files and where there come out that is a code that work that shows not just you there are mean people who put codes on this site that don't work first time have so mean errors in them

Excellent work!!! Thank you for your repo.

Is there a gate (which is denoted by G in formula (3) in the paper) in this smoothness loss?

Hi,
I'm sorry to disturb you again.
I am collecting saliency maps for other methods.
Due to some omissions, I hope you could provide the result maps you collected as a supplement.
When the work is completed, I will introduce your contribution。
Thanks a lot.

Hi, thanks for sharing the code. Could you please provide the VGG pretrained model for training the network from scratch?

I cannot share to the google yunpan in my account.Why?

Thank you for the great work!

In Fig.4 of the paper, you indicate that the intensity image of the input image is used during the model training, yet there are limited details concerning how the intensity image is generated based on what mechanism. Also, the scripts of the provided convert_rgb2gray.m are not quite self-explainable.

Could you please provide more details on the image intensity value generation logic?
Thank you very much!

I wonder how much of an impact training with different scribble datasets would have.

Hi,
Thanks for sharing the code. I can train and test your models now. But I cannot evaluate the results. Could you please provide the evaluation code for generating the metric numbers in Table 1?
Thanks a lot!

Hi, Jing,
Thank you for sharing the wonderful work.
I have tested the trained model you uploaded. But the results are different from the paper. Is the uploaded model your final result of your paper?

Hi, Jing,

I am interested in your novel metric. Would it be possible to release the metric code first?

Thx,
Xin

No datasets were found under this folder

Hi， thanks for your contribution, but when I run the train.py, it has a error:[Errno 2] No such file or directory: '/home/jing-zhang/jing_file/vgg16-397923af.pth', when I change it ,it will has an error: 'odict_keys' object is not subscriptable

Hi! Thank you so much for your work. May I ask when will you release the code?

I cant undstand the problem:TypeError: 'odict_keys' object is not subscriptable.
In the vgg.py self.conv1.conv1_1.weight.data.copy_(pre_train[keys[0]])

Hello, I want to know how do you get the edge by sobel rather than RCF? Could you show the related code?:smile: