Yi Wang

When using the pretrained model on Places2, please set the image and mask shapes by `--img_shapes 256,512,3 --mask_shapes 256,256`. About the image resolution in Places2, we use the [data in...

Sure. If you wanna freeze some specific layers in the generator, you can remove them by their names in [g_vars](https://github.com/shepnerd/outpainting_srn/blob/c0cc6fc931b9174d2c6b25463576d57329091089/train.py#L27), then these parameters will not be updated in the following...

> Hi Yi Wang. What applications do you think this technology has in real life? For example, what kind of practical problems can be solved? Extending images or videos naturally...

> I am training from the newly selected data set. To what extent can the first stage of training be identified as convergence？thanks for your help We can ensure the...

> > > Can you send me all the loss curves of your training at that time? thanks. I will search my server for these data and get back to...

> > >  > > Is the loss function image of this discriminator normal? I trained from scratch with a newly selected dataset. At least the loss tendency of...

The posted results did seem problematic. Did the first stage training converge? Could you share the training loss trend of both first and second stage?

It seems you need to training the first stage longer since you conduct natural scenes inpainting. Besides, you may reverse the channel order of the shown images. It appears in...

Some functions and losses do not support parallel running due to the design flaws in that time. We hope to improve them soon but we are with other projects now.

We use 8 NVIDIA 2080ti GPUs in experiments. Batch size in training affects final generation performance. Testing with one GPU is fine.