It’s difficult today to miss the meteoric rise of artificial intelligence in the fields of image and video. But between often limited online platforms and complex tools to use, finding a good balance remains a challenge. This training dedicated to ComfyUI offers a different approach: regaining total control of your creations, locally, without depending on a third-party service.
ComfyUI is gradually establishing itself as a reference for demanding creatives. More technical, but also infinitely more flexible, it allows you to build your own workflows and fully exploit the power of recent open source models. This is therefore the purpose of this training presented in partnership with Tuto.com.
Who is this video tutorial for?
This AI training is primarily aimed at a public already aware of the generation of images by AI. It is particularly suitable for:
- photographers and videographers curious about integrating AI into their workflow,
- to creatives who wish to go beyond the limits of online tools,
- advanced users wishing to customize their generation pipelines,
- to those who favor confidentiality and local work.
On the other hand, complete beginners might find the approach more demanding than with turnkey solutions.
The training program
Lasting approximately 9 hours, the training is organized into large thematic blocks and here is an overview of the program:
Getting started with the tool
- Installation et interface
- Création et organisation de workflows
Génération d’images
- Prompting assisté (ChatGPT, Gemini, Ollama)
- Image vers prompt et workflows automatisés
- Comparison of recent models
Customizing renderings
- Using Loras
- Adjusting visual styles
Amélioration et édition d’image
- Upscale and automatic correction
- Advanced editing with Qwen Edit
- Inpainting et extension d’image
Control and Styles
- Using ControlNet (Canny, Depth)
- Transferring and Combining Styles
Video creation with AI
- Génération vidéo (texte ou image)
- Motion control and animation
Finition vidéo
- Upscale en 1080p
- Interpolation for smooth playback
Many source files are also provided with the training so that you can practice on the same examples.
Create a prompt from an image
This video chapter offered to you allows you to concretely understand how an advanced workflow works in ComfyUI. The idea is simple: start from an existing image to automatically generate a usable prompt. To do this, the author uses an AI model capable of analyzing visual content (objects, atmosphere, composition…), then translating it into textual instructions.
