AI is finally controllable! …with a stick [FREE ComfyUI Workflow + Tutorial]

Опубликовано: 22 Декабрь 2025
на канале: Mickmumpitz
38,842
2.4k

Control AI video like a puppet master! This free workflow lets you guide AI characters with a stick, toy, or 3D layout animation.

If you like my work, please consider supporting me on Patreon:   / mickmumpitz  
Follow me on Twitter:   / mickmumpitz  

You can find the FREE WORKFLOWS here:   / new-video-free-i-146074703  
INSTALLATION GUIDE ComfyUI: https://mickmumpitz.ai/guides/install...
WORKFLOW GUIDE: https://mickmumpitz.ai/guides/ai-vide...
You can find the EXAMPLE FILES here: [Coming Soon]
We also created ready-to-use RunPod* templates for both workflows: https://console.runpod.io/deploy?temp...

TOOLS:
Time-to-Move (TTM) https://time-to-move.github.io/
Wan 2.2 Video Model https://github.com/Wan-Video/Wan2.2
SAM 3 (Meta) https://ai.meta.com/sam3/
Qwen Image Edit https://qwen.ai/blog?id=a6f4837771446...
ComfyUI https://www.comfy.org/
Wan Animate https://humanaigc.github.io/wan-animate/
AI Toolkit (LoRA Training) https://github.com/ostris/ai-toolkit
RunPod https://runpod.io?ref=mgaruyku *

CHAPTERS:
00:00 Intro
01:00 How it works
02:20 Preparation
05:22 Time to move!
07:23 De-Aging myself
08:55 Final short film

Controllable AI video that actually works. This tutorial demonstrates Time-to-Move (TTM), a training-free method for precise object and character animation in AI video. Move toys, cutouts, or 3D previs through your scene, and the workflow automatically masks, removes rigs, and generates clean animation with proper shadows and reflections.

We also cover Wan Animate for transforming yourself into different characters and de-aging, plus LoRA training to maintain consistent faces across multiple shots. Watch us create a complete AI short film using these techniques—perfect for AI filmmakers, animators, and anyone tired of fighting unpredictable AI motion.

*Affilliate links: If you sign up to RunPod using our link and add $10 to your account, we both get bonus credits. (More info: https://docs.runpod.io/references/ref...)