Strange output with red noise point

[AlexCHENSiyu]

I can get this good output yesterday.
but today when I ran it again, I get this strange output:

Have you ever encountered this problem before? Or you can give me some suggestions?

[AlexCHENSiyu]

ok, probably I find the reason. BTW , this part should be written like this:

add ['arr_0'] to it. otherwise cannot run.

[ivanstepanovftw]

I had this when I did not clamp styled image:

artistic-style-transfer-for-videos/main_video_join.py

Line 57 in b42b79d

styled.data = styled.data.clamp_(min_vals[:, None, None], max_vals[:, None, None])

Similar as it is performed in the closure:

artistic-style-transfer-for-videos/main_video.py

Line 342 in b42b79d

styled.data = styled.data.clamp_(min_vals[:, None, None], max_vals[:, None, None])

I actually thinking about adding loss for out of bounds pixels, so the optimizer would not be focused on these pixels.

[AlexCHENSiyu]

yes, right after I posted the issue, I found this mistake of mine.
I read the paper carefully and I realize that long-term temporal consistency requires heavy-duty computation. It might be hard for me to achieve that.

[AlexCHENSiyu]

BTW, do you know any comparative standard to measure the quality of style-transferred images between different pre-train model?

[AlexCHENSiyu]

and any reason why you choose efficientnet-b0 as the best model

[ivanstepanovftw]

BTW, do you know any comparative standard to measure the quality of style-transferred images between different pre-train model?

Unfortunately, no.

and any reason why you choose efficientnet-b0 as the best model

I am not sure if it is best for style transfer. EfficientNet chosen as it is efficient, since I do not have local GPU.

Also, in the paper "CNN Filter DB: An Empirical Investigation of Trained Convolutional Filters" (arxiv:2203.15331v2), they say that "learned filters do not significantly differ across models trained for various tasks, except for extreme outliers such as GAN-Discriminators."