Text2Video TD: Operator for AI-Generated Videos
Added 2023-05-03 22:16:03 +0000 UTCHey friends!
I am super excited to share Text2Video, a TouchDesigner operator designed to generate AI-generated videos using text prompts or input videos or images !
[ Project file includes a in-network GUI for easy control as well as the prompts used to create these videos !! AVAILABLE ON DISCORD - or if thats not your thing, shoot me a message on here or instagram and I can share a link for Patreons. ]
[Nvidia / CUDA cards are required as far as I know.]
This operator extends the functionality of the API in the Text2Video extension for Automatic1111, providing an easy-to-use interface within Touchdesigner.
By using Text2Video, you can generate short videos based on text prompts, adjust video parameters, and have full control over the video generation process.
Some of the key features that the TD tool adds to the automatic1111 extension are:
- Browse past video outputs / settings quickly and easily
- Track settings with a settings_vid.txt file created for each generation
- One click to upscale the current display video to remove watermarks/increase resolution
- Built-in Touchdesigner extension for Python-based control and further customization
To get started with Text2Video, you'll need Automatic1111 with the Text2Video extension installed, which can be found here: https://github.com/deforum-art/sd-webui-text2video. Automatic1111 Extension Features:
- Generate videos based on text prompts and various parameters
- Vid2vid mode - Generate video based on any input video from disk
- Inpainting mode - Insert an image from disk at one or multiple points in the video generation
I have included a more detailed readme built into the op [ html readme ! ].
Check the About tab of the operator or dive inside to see it.
I appreciate your continued support, and look forward to seeing what you create !
I have been a little quiet on Patreon but the development has continued. More coming soon !!! :)
Comments
Hi, How can I access the file?
Chung-Ta Huang
2024-11-03 05:20:36 +0000 UTCThanks for reply here and also on discord. I was able to get automatic1111 to connect via the code you shared (it worked for the still image creation…but I couldn’t get to a point where anything’s moving/animations). I will try again later in the week. Appreciate your sharing these tools and look fwd to spending more time with them and your resources when I get a break. Cheers.
Kathleen Judge
2023-05-09 20:13:24 +0000 UTCApologies for missing this. I have most of my resources on discord at the moment, but I am in the process of organizing my various tools and creating better resources/readmes for all of the things. I am trying to make things easier to access for sure.
dotsimulate
2023-05-09 05:21:14 +0000 UTCI guess the part I can't seem to figure out how to do is to add an 'api flg'/ I already have automatic111 and SD locally installed and use in a regular way. I'm just not sure what this instruction means: "Start A1111 with api flag" Will also post this question in the discord.
Kathleen Judge
2023-05-09 04:43:12 +0000 UTCthank you. I will check that out. I was thinking maybe I'd missed some technique or input/button within the dotSimulate files that activates automatic1111 but I guess not. Appreciate the link.
Kathleen Judge
2023-05-05 16:19:45 +0000 UTCI do know there's one on github: https://github.com/olegchomp/TDDiffusionAPI
Ken Hill
2023-05-05 14:54:59 +0000 UTCIs there a a tutorial for connecting TD to automatic111? I have Stable diffusion/deforum/automatic111 running locally. I just don't know how to connect it into the TD patch?
Kathleen Judge
2023-05-05 00:05:00 +0000 UTC
