Picsart-AI-Research / Text2Video-Zero

[ICCV 2023 Oral] Text-to-Image Diffusion Models are Zero-Shot Video Generators
https://text2video-zero.github.io/
Other
3.91k stars 336 forks source link

About the Background Smoothing #68

Open FatLong666 opened 9 months ago

FatLong666 commented 9 months ago

Hello, when I run the project to generate a video, I want to use the Backgroud-Smoothing to make a compression between the different video. But I got some error about the Backgroung-Smoothing.

File "D:\pycharmprojects\Text2Video-Zero\text_to_video_pipeline.py", line 428, in call m_f = torch.tensor(self.sod_model.process_data( AttributeError: 'TextToVideoPipeline' object has no attribute 'sod_model'

When I spent some time to figure it out, I found that the Text_to_video_pipeline.py doesn't define the self.sod_model. It seem the project also have not provide more information or fully code about the Background-Smoothing.

I hope U can help me.

Best, regard!