ComfyUI Generative AI Production for VFX – Course Bundle

ComfyUI Generative AI Production for VFX – Course Bundle

$299.00

ComfyUI Generative AI Production for VFX – Course Bundle

This bundle combines AIF201 and AIF202 into a single learning pathway, taking you from foundational generative AI workflows to advanced video production techniques with ComfyUI. Together, these courses offer a practical, production-focused understanding of generative AI in modern VFX pipelines. You will begin with core workflows for image generation, texturing, and asset creation, then advance to video generation, camera control, and animation workflows integrated with Maya and Nuke.

AIF201: Generative AI for VFX with ComfyUI & InvokeAI

Taught by Eran Dinur
Intermediate Level Course

This course provides VFX and visualization artists with a practical understanding of the Stable Diffusion generative process using two versatile node-based applications: ComfyUI and InvokeAI. You’ll learn how to build custom workflows for tasks such as generating seamless textures, creating matte painting elements, and producing fully textured 3D assets. The course covers core components of the generative pipeline, including models, ControlNets, preprocessors, IP adapters, and techniques for inpainting and upscaling. By the end of the course, you will be able to design and build workflows from scratch, using generative AI as a flexible tool within your VFX pipeline. No prior knowledge of generative AI is required. Basic familiarity with Nuke, Maya, or Photoshop is recommended.

**Update Note: A recent ComfyUI update is causing a malfunction in the Hunyuan 3D wrapper used in the course. There is an easy fix for this, detailed in the downloadable PDF ‘AIF201 – Fix for Hunyuan 3D’.

AIF202: ComfyUI Generative Video Production for VFX

Taught by Eran Dinur
Intermediate Level Course

Building on AIF201, this course focuses on video generation workflows in ComfyUI, with an emphasis on camera control, animation, and integration with traditional 3D and compositing tools. You will learn to work with WAN models for video diffusion, including single- and dual-model (low- and high-noise) setups, and to develop techniques for smooth, cinematic camera motion. The course also covers transferring camera and character animation from Maya into ComfyUI using ControlNets and OpenPose.Advanced workflows include using Wan Wrapper tools, Uni3C, and TimeToMove for camera and animation control, as well as practical approaches to video upscaling, temporal enhancement, and finishing in Nuke. A complete, production-ready approach to generative video within a VFX pipeline.

Instructor: Eran Dinur

Eran Dinur is a VFX Supervisor whose credits include Marty Supreme, Hereditary, The Wolf of Wall Street, and Uncut Gems. He is also the author of The Filmmaker’s Guide to Visual Effects and The Complete Guide to Photorealism.

 

AIF201: Generative AI for VFX with ComfyUI & InvokeAI

Class 1: Introduction and Setup

We discuss the advantages (and challenges) of working with ComfyUI and InvokeAI and compare the two applications. We then set up a comprehensive environment for Comfy UI that facilitates the installation of complex custom nodes and their dependencies.

Class 2: Essential SD Building Blocks

We build our first generative workflow in ComfyUI, exploring the fundamental components of the Stable diffusion process (Checkpoint models, clips, samplers, latent noise, VAE decoder) and learning the functions of essential controls like CFG, seed and steps.

Class 3: Seamless Textures Tool: Part 1

We start putting together a dedicated workflow that can generate seamlessly tileable textures for 3D. We learn the function of various nodes in InvokeAI and how to use split prompts.

Class 4: Seamless Textures Tool: Part 2

We continue developing our texture generation tool by adding various controlNet nodes and preprocessors, allowing us to use other images as guides. We also look at the unique features of the SDXL checkpoint model.

Class 5: Using CG Elements as Guidance: Part 1

We jump between Maya, Nuke and ComfyUI to create a modern villa environment. We use the beauty, Z depth and normal passes to accurately guide the AI generation toward the desired composition and layout.

Class 6: Using CG Elements as Guidance: Part 2

We continue with the villa project, using Cryptomattes for regional prompting and IP adapters for lighting, style and mood.

Class 7: Generating 3D Assets Part 1

We begin building a comprehensive workflow for generating fully textured 3D models using the HunYuan model and wrapper nodes. We go over installation and dependencies and then build the image-to-mesh and de-lighting sections.

Class 8: Generating 3D Models Part 2

We continue our step-by-step workflow for building with AI, including UV mapping, multi-camera setup, texture projections, unwrapping, and final export of the textured mesh, while exploring tools such as the multi-view renderer and the CV2 Inpaint.

Class 9: Inpainting with InvokeAI

In this class, we step back from node-based workflows and explore InvokeAI’s excellent layer-based inpainting interface, examining different techniques for assisting with matte painting and compositing tasks.

Class 10: Segmentation and Upscaling Techniques

We wrap up the course with a look at segmentation and image-to-language with the Florence model, then examine different techniques and tools for upscaling.

 

AIF202: Advanced ComfyUI – Generative Video Production for VFX

Class 1: Introduction to Video Generation and System Setup

We introduce video generation workflows in ComfyUI, including how they differ from still image generation. We examine VRAM consumption, memory swapping, and strategies for working beyond local hardware limitations using RunPod and Comfy Cloud. We also break down WAN model classes and naming conventions to establish a solid technical foundation.

Class 2: From Still Images to Video Workflows

We convert a standard checkpoint-based workflow into a basic video diffusion pipeline. This includes building both text-to-video and image-to-video workflows using the IT2V 5B model, while understanding the transition from still to temporal generation.

Class 3: Dual-Model Workflows and Camera Motion Control

We build an effective high–low noise two-model workflow to improve generation stability and quality. We also introduce Wan Camera Control, enabling smooth and controlled cinematic camera motion within generated sequences.

Class 4: Video ControlNet – Scene Setup and Rendering (Part 1)

We begin integrating CG workflows by setting up a scene and camera animation in Maya and rendering the required elements. These renders will be used to guide video generation through ControlNet.

Class 5: Video ControlNet – Camera Transfer (Part 2)

We transfer complex camera motion from CG into generative workflows, using ControlNet to align AI-generated video with precise camera animation created in Maya.

Class 6: Advanced Camera Control with Wan Wrapper and Uni3C

We create Wan Wrapper workflows and explore Uni3C for fully independent camera control, enabling more flexible, production-ready camera systems.

Class 7: Character Animation Transfer with OpenPose

We transfer character animation from Maya into ComfyUI using OpenPose, enabling the integration of animated performances into generative video outputs.

Class 8: Advanced 2D-to-3D Camera Motion Techniques

We develop advanced camera workflows using TimeToMove and Wan Wrapper, enabling more complex 2D-to-3D camera motion and improved spatial consistency.

Class 9: Video Enhancement and Finishing Techniques

We explore TimeToMove animation tools alongside techniques for effective video upscaling, temporal enhancement, and motion blur. We also integrate ComfyUI outputs into Nuke for final enhancement and finishing.