DoubleXING

> @Doubiiu How long did it take? For me, it takes almost two weeks to train the code. I wonder if this is common situation Ahhhh Actually I haven't started...

> > @Doubiiu How long did it take? For me, it takes almost two weeks to train the code. I wonder if this is common situation > > Hello! May...

Hi Thanks for your interest! I am not sure how the model performs on the arbitrary resolution inputs. You can directly try it using both 512 model and 1024 model,...

Hi, what is your exact resolution for inference? To my best knowledge, the height and width should be divided by 32 (I think this is the reason of your bug)....

Hi Thanks for your interest! We are currently working diligently to improve the model's stability at a resolution of 320x512, which includes image animation and generative frame interpolation. We will...

😭Sorry for the late reply, we have released the generative frame interpolation and looping video generation model & their corresponding gradio demo. Have a try if you are still interested...

Hi I have updated the `i2v_test.py` file to fix the corresponding bug. Sorry for the incovenience caused. Now It will not redownload the model if it already exists.

Hi Theis is caused by loading the DynamiCrafter256 model checkpoint into the DynamiCrafter1024 model. (because of the bug in checkpoint-downloading script `i2v_test.py` and now I have fixed it.). You can...

Hi. Sorry for the late reply. I generated 2048 samples (use frame_stride=3) using the 1st frame of 2048 randomly selected videos in MSR-VTT. When computing FVD, please also use frame_stride=3...

Which model did you try? For 512 model, I have added `perframe_ae=True` into the config as pointed out by Issue #18, where there should be no OOM problem for the...