SystemErrorWang

@pdsing To cartoonize videos, we at first split them into frames, process each frame with the model, and finally concatenate them back to a video.

In case of noise, I guess it was because of the guided filter layer. you can try to train a model without the guided filter in the generator and then...

@itomorrower08 This may be because of environment conflict. In my own experiment, tensorflow=1.12.0/1.13.0, scikit-image==0.14.5 and joblib==0.13.0.

@itomorrower08 https://drive.google.com/open?id=1JfJzJbNjAWBIHGm9mc_R9dXv7DAw3tZc you can download this file here

@itomorrower08 scenery photo：cyclegan training set；scenery cartoon：miyazaki hayao 3617 + hosoda mamoru 5107 + shinkai makoto 5891；face photo：ffhq 00000-10000；face cartoon：pa works 5000 + kyoto animation 5000

@Xinxiang7 we used the dataset provided in CycleGAN repository for scenery dataset. You can select the first 5000 images or use all 6227 images, this will only cause minor differeces.

@Xinxiang7 yes, it contains a wide variety of scenes and topics. for details of the dataset, please download and check it.

@alexanderpoplavsky you can download the training set here, but it will be deleted if there're any copy rights issus https://drive.google.com/file/d/10SGv_kbYhVLIC2hLlz2GBkHGAo0nec-3/view?usp=sharing

@Simension well thank you for asking, in fact i do not care about this, the only problem is that this cartoon filter may not be good enough to help you...

load the weight and save it again use tf functions, you can get the .meta file