Courses Architecture The Diffusion Architect: Stable Diffusion for Practical Workflows

The Diffusion Architect: Stable Diffusion for Practical Workflows

The Diffusion Architect teaches practical AI tools for architectural design, covering fundamentals, prompting, and custom workflows.

00:00

00:00

/

02:49
- 2x
- 1.5x
- Normal
- 0.75x
- 0.5x

Overview
Content
Instructors
Reviews

Date:

Jul 27, 28, 2024

Duration:

8 Hours

Difficulty:

Beginner

Language:

English

Certificate:

Yes

Registration:

€161.50

Tutor:

Ismail Seleit

Members:

137.28 EUR (27.75%) discount

Recordings:

Available Indefinitely

Gift The Course

Introduction of the workshop:

This workshop’s main objective is to take you from zero to hero for the practical use of AI tools in the architecture design workflow. These skills are a must-have in this ever-changing landscape of the creative industry and will help you focus more on what is essential while designing. The workshop will revolve around image diffusion models and go through crucial elements and definitions necessary to get value out of these models.The focus will be on open-source Stable Diffusion combined with one of the most important extensions for practical use, ControlNet. We will cover the basics and progress through advanced workflows tailored to specific design goals. The participants will leave the workshop able to design their project-specific workflows and tackle design problems using AI as actual design assistants.

What You’ll Learn?

Understand the fundamental principles of Stable Diffusion and its application in architectural design.

Learn effective prompting strategies and advanced techniques for generating architectural designs.

Gain proficiency in using ControlNet for enhanced precision in design workflows.

Be capable of creating custom workflows tailored to specific design objectives.

Utilize AI tools as design assistants to streamline the design process and boost creativity.

Have access to workshop materials and exercises post-completion for further experimentation and development.

The Diffusion Architect: Stable Diffusion for Practical Workflows

Methodology:

The workshop consists of two-day structured exercises, progressively building participants’ knowledge of basic principles to advanced techniques. We will start with theoretical foundations and hands-on practice, ensuring participants can navigate and configure Stable Diffusion settings.

The sessions will cover effective prompting strategies, ControlNet integration, and custom workflow creation. Each exercise will be followed by a collective assessment of the outputs, enabling participants to refine their techniques. This immersive approach ensures that participants can directly apply their knowledge to architectural design projects.

In The Diffusion Architect workshop, we will explore the practical application of Stable Diffusion in the architecture design workflow. We will introduce the technology and fundamental concepts, followed by hands-on practice configuring and navigating Stable Diffusion settings.

The workshop will then delve into effective prompting strategies, the use of advanced features, and the integration of the ControlNet extension for precision control. By the end of the workshop, participants will have the skills to create custom workflows for their specific design goals and apply AI tools effectively in their projects.

What You’ll Use?

ComfyUI

Stable Diffusion

Highlights:

Content

Introduction to the workshop – PAACADEMY Updates (08:19)

Introduction & Updates

Lecture – Stable Diffusion Overview (01:14:24)

Stable Diffusion Basics – Image Generation, Fine-Tuning, and Prompting

In this lecture, Ismail covers stable diffusion, including image generation from text prompts, the impact of fine-tuned models, and the role of prompts, seeds, and interfaces like Automatic1111 and ComfyUI.

Generating Images with Automatic 1111 (51:08)

Image Generation with Automatic1111: Prompts, CFG Scale, Seeds, and Loras

This session covers generating images with Automatic 1111, adjusting prompts and settings like CFG scale, and using seeds and Loras.

Using ControlNet with Stable Diffusion (47:25)

ControlNet in Stable Diffusion: Image Guidance and Installation

This session covers using ControlNet in stable diffusion to guide image generation with input images, Loras for architectural design, and how to install and adjust ControlNet for optimal results.

Mastering Stable Diffusion with ControlNet (01:09:03)

Stable Diffusion and ControlNet for Architecture: Inputs and Troubleshooting

This session explores using stable diffusion and ControlNet for architectural design, including applying ControlNet with various inputs, adjusting parameters, using Loras for style, and troubleshooting common issues.

Introduction to Day 2 (02:26)

Images from Sketches with Stable Diffusion and Control Net (01:02:52)

Stable Diffusion and ControlNet: Generating Images from Sketches

This session explores using Stable Diffusion and ControlNet to generate images from sketches, focusing on model selection, invert pre-processing, and Control Weight parameter effects.

Enhancing Image Generation with IP Adapters in Stable Diffusion (49:50)

Image Prompts in Stable Diffusion: Style Transfer and Conditioning

This session explores using image prompts in Stable Diffusion with IP adapter control nets for style transfer, graphical consistency, and conditioning generation based on reference images and 3D massing models, emphasizing experimentation and iteration.

Stable Diffusion in Architectural Workflows (57:03)

Stable Diffusion for Architects: Image Generation and Workflows

This session highlights using Stable Diffusion for architects, covering techniques such as image-to-image generation, inpainting with control nets, and prompt scheduling to create custom workflows while emphasizing compatibility among various Stable Diffusion tools.

Overview of Stable Diffusion Tools for Image Generation (01:02:36)

Stable Diffusion Tools: Control, Simplicity, and Speed

This session reviews tools for generating images with Stable Diffusion, focusing on their control, simplicity, and speed trade-offs.

Quick AI Discussion (18:04)

Hamid Hassanzadeh and Ismail Seleit: Future of Image Generation

A conversation between PA’s Founder Hamid Hassanzadeh and Ismail Seleit about Stable Diffusion and the future of image generation tools

Important Notes:

Press here for a detailed list of all the software/Plug-ins needed.

Instructors

Ismail Seleit

12 Followers

0 Following

Biography

Ismail Seleit is a distinguished architect at Foster + Partners, renowned for his expertise in computational design and BIM (Building Information Modeling). His work encompasses various projects, from design competitions to complex building realizations. Ismail is dedicated to integrating advanced AI tools, such as Stable Diffusion and ControlNet, into architectural workflows, pushing the boundaries of design innovation. Beyond architecture, he is an accomplished ambient-electronic music producer, creating soundtracks for architectural films and various independent projects, showcasing his multifaceted creativity and technical skills.