kris-singh / deep-unsupervised-saliency-detection Goto Github PK

I have read your paper and your code, I have some questing to ask your help ?

1. I'm confusing about the noisy saliency maps, I want to know how to generate noisy saliency maps during train.
2. in your code, there is a sal_label in your dataloder.py, so whta's the sal_label?
3. the method described in the paper is unsupervised, but in the train precedure, labels are used, so I'm really confusing.
   can you explain these above questions? thank you!

Hi, I have a small dataset (300 training images). The model performs ok during round 1 of training, but when introducing the noise module, the prediction loss returns to where it started in round 1 and does not improve. The noise loss does not decrease over time either. Any ideas? I'm using two sources of noisy ground truth.

Dear author, is it convenient for you to provide your trained model and testing script? Thank you very much.

Thanks for your implmentation ! Recently, I am interested in the unsupervised learning, this implmentation can help me. Here, I don not know the meaning of "Exp Name FULL". Can you explain the Full Training?

Thanks very much for your code.
But I am a little bit confused of the dimension of noisy_gt.
I use opencv to generate noisy ground truth and the dimension is (256,256),is that the true dimension?
And I noticed that the input image x is repeated by NUM_MAPS,I don't understand the meaning of it.

hi, thanks for you work. i tried your code, but came into with the error: No such file or directory: '/data/workspaces/krish/sd/msrab_hkuis/small.lst'.
so how to get or create the data? can you update part of some which can be directly used or a final official version?

I would be very grateful if you could describe how to use your code. For example, what is the form of train.lst ? Thank you very much.