Instructions to use CiaraRowles/TemporalNet with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use CiaraRowles/TemporalNet with Diffusers:
pip install -U diffusers transformers accelerate
from diffusers import ControlNetModel, StableDiffusionControlNetPipeline controlnet = ControlNetModel.from_pretrained("CiaraRowles/TemporalNet") pipe = StableDiffusionControlNetPipeline.from_pretrained( "runwayml/stable-diffusion-v1-5", controlnet=controlnet ) - Notebooks
- Google Colab
- Kaggle
Is this corret?
In line #95 of temporalvideo.py it says:
send_request(x_path,y_folder, y_paths[0])
where x_path is the reference image path and y_paths[0] is the first image to be processed.
But in line #106 it says:
send_request(y_paths[i], y_folder, temp_image_path)
where y_paths[i] should be the image to be processed and temp_image_path is the last image to be used as a reference.
Am I correct?
O.O oh nooo, oops, yeah you're right, good spot.
That's interesting as i'd assumed it would have broken the whole thing if wrote it like that by accident, since the hed is always inputted with the last frame, not the original image, oh dear.
let me update that.
Can't thank you enough, this actually explains quite a few oddities. What i get for being lazy and letting chatgpt do it xD i'm a c# Unity developer by trade.
well 60-70% of it roughly.