results on AISTD
hieulem opened this issue · comments
Hi there, congrats on the great paper!
I wonder if you can provide the results of the model trained on the AISTD dataset where the testing images are kept as they were provided in the original ISTD dataset?
Hi, thanks for your attention.
We report the results on ISTD because we want to compare other methods fairly. Actually, we are running some experiments on a similar dataset like AISTD (using the pre-processing technique in https://arxiv.org/pdf/1805.04635.pdf). The results are still good.
I will try to run the model on AISTD. but it requires some time.
Oh, thank you!
Do you have the results of the model trained on ISTD+ and tested on the original ISTD?
Hi again :).
I used your code to produce some results for the ISTD dataset. I observed that the output images look a bit darker than the input images. Is it normal? Is it okay to use them for comparison?
I am sorry to reply late.
There are some really busy days and I think the additional experiments will be finished in December.
I didn't observe this phenomenon and can you share some examples with me?
Overall, I think you can use these results for comparison if you are eager.