search

Weifeng-Chen / control-a-video

Official Implementation of "Control-A-Video: Controllable Text-to-Video Generation with Diffusion Models"
GNU General Public License v3.0
359 stars 26 forks source link

where to put the HED pth file? #24

Closed G-force78 closed 1 year ago

G-force78 commented 1 year ago

I cant find the instructions on the readme, neither can I see where the depth model ends up.

Weifeng-Chen commented 1 year ago

I cant find the instructions on the readme, neither can I see where the depth model ends up.

you can download from here: https://huggingface.co/wf-genius/controlavideo-hed/blob/main/hed-network.pth

G-force78 commented 1 year ago

Yes have seen that thanks, I have now found out the path needs to be changed in inference.py to the working directory or wherever you downloaded it to. eg

_``` if control_mode == 'depth': annotator_model = DPTForDepthEstimation.from_pretrained("Intel/dpt-hybrid-midas").to("cuda") elif control_mode == 'canny': annotator_model = None elif control_mode == 'hed':

firstly download from https://huggingface.co/wf-genius/controlavideo-hed/resolve/main/hed-network.pth

annotator_model = HEDNetwork('/content/control-a-video/model/annotator/hed-network.pth').to("cuda")