Course Content
11 Lessons
Loading
PixelSpace: Intro to 3D Generative AI focuses on automating 3D scene creation using AI models and Unreal Engine.
1 course5.0
Recent image and video generation advancements have revolutionized content creation, offering unprecedented possibilities. While we rapidly approach seamless interaction with scenes and videos through text inputs, 3D scene generation remains a fertile ground for exploration. Over the past two years, significant progress has been made in 3D and 4D object generation, mainly through multiview diffusion models. More recently, the application of consistent video generation as a proxy for 3D spatial representation has emerged, paving the way for comprehensive 3D scene generation.
These developments present a unique opportunity to harness deep learning algorithms, allowing us to envision and create immersive 3D scenes, thereby unlocking a new dimension of creative expression.This workshop will explore the cutting-edge techniques of multiview diffusion models conditioned on camera paths for 3D scene generation.
We will dive into the latest methods for representing 3D scenes and geometry and investigate how current research in 3D generative AI leverages pre-trained image and video models for a 3D generation. Participants will learn how to use text and image inputs to generate scenes, which will then be imported into Unreal Engine for post-production and to create a concept reel.
This workshop focuses on the automation of 3D scene generation from text and images. Through theoretical discussions and practical exercises, participants will explore the capabilities of 3D generative AI and its application in fast-tracking design exploration. Key topics include:
Introduction to Multiview Diffusion Models:
3D Scene and Geometry Representation:
Leveraging Pretrained Models:
Text and Image Inputs for Scene Generation:
Integration with Unreal Engine:
Exploring 3D Generative AI : Understand and utilize advanced AI algorithms for scene generation.
Scene Creation : Learn to automate the process of 3D scene creation using text and image inputs.
Interactive Design : Develop skills in creating interactive and immersive experiences in Unreal Engine.
Creative Exploration : Foster innovation and creativity in architectural design through AI tools.
Inspirational Inputs:
Scene Generation:
Interactive Experience Creation:
Project Showcase:
Course Content
Daniel Escobar is a creative designer and 3D generative AI researcher passionate about pushing the boundaries of creative workflows and enhancing the creative process. His primary focus lies in exploring creative applications for 3D generative neural networks. He cofounded OLA, which focuses on developing projects through various media, including architecture. Daniel has been published at Caadria, Acadia, Archdaily, and Dezeen and has taught multiple workshops exploring the intersection of AI in creative applications. He also founded Diffusion Architecture, a platform that records and disseminates the creative use of AI in architecture and design.
Very interesting course.