Myung-Joon Kwon

Thanks. Then no ImageNet weights for BiSeNet v2? Plus, could you tell me how to combine encoders with heads?

Thank you :D I have more questions. 1) Do we have to normalize the inputs when using ImageNet pretrained models? 2) When I use BiSeNet (either v1 or v2), this...

Hi, It seems that the 'input' directory is empty. Put images there and run infer.py. Thanks

Glad it works now!

Email me.

Update: So many people are asking me for the FantasticReality dataset. If you need this for reproducing CAT-Net results, email me via **[email protected]** Please don't send emails to any other...

I just discovered that some emails have been ending up in the spam folder. Please avoid using domains specific to certain countries. I recommend using your school or work email...

The datasets used for training can be found in the paper or the code: [link](https://github.com/mjkwon2021/CAT-Net/blob/dbdd2305dcfe33f371eb7de2c40c3fba554183d2/Splicing/data/data_core.py#L29) I cannot say for sure why your model cannot train well because I have no...

Probably the learning rate might be too large. The learning rate decreases substantially as the training progresses. When you try to fine-tune the pretrained model, the learning rate should not...

Hi, I cannot infer errors from this one sentence. You should debug carefully. Follow your input image and find what makes nan.