How can I use feature Multi-Text Conditioned Long Video Generation?

[kadirnar]

Can you update the Colab demo? Can you share a tutorial on how to use it?

I want to make this video.
https://github.com/G-U-N/Gen-L-Video#multi-text-conditioned-long-video-generation

[G-U-N]

Hi, thanks for your attention. I am actually considering doing that. But I am currently busy with other projects (e.g., my bachelor thesis) and making more demos.

The core idea of Gen-L-Video is very simple and straightforward. Here's what I am planning to do in the future.

1. A tutorial to demonstrate the core idea of Gen-L-Video.
2. Release more demos and their corresponding scripts.

[kadirnar]

Thank you very much for this great demo. I really liked the anime video you made. Can you share the sample code here to do just that?

Or what Python files should I examine to do this?

[G-U-N]

That demo was generated with one-shot-tuning-mix, which has not been uploaded or thoroughly checked to ensure everything is fine. I will upload it as soon as possible.

[Ainaemaet]

Very exciting.
Currently testing SD/CN and have gotten some interesting results, but this looks quite promising.
Second the idea of a tutorial - I'm sure it won't be too difficult to get it all going but if there is any expectation for some of the newer people just figuring this stuff out it will likely be a nightmare for them lol.

Thanks, looking forwards to getting that one-shot-tuning-mix available!

[shagharabbani]

Hi, Can you guide me on why do I get an error when I change the video for tuning-free-mix? What are assert len(prompt) == video_length//clip_length.

[wang-zm18]

Nice work! When will you release the instructions for reproduce https://github.com/G-U-N/Gen-L-Video#multi-text-conditioned-long-video-generation? Thank you in advance!