Holodeck: Language Guided Generation of 3D Embodied AI Environments

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Yue, Sun, Fan-Yun, Weihs, Luca, VanderBilt, Eli, Herrasti, Alvaro, Han, Winson, Wu, Jiajun, Haber, Nick, Krishna, Ranjay, Liu, Lingjie, Callison-Burch, Chris, Yatskar, Mark, Kembhavi, Aniruddha, Clark, Christopher
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911849459482624
author Yang, Yue
Sun, Fan-Yun
Weihs, Luca
VanderBilt, Eli
Herrasti, Alvaro
Han, Winson
Wu, Jiajun
Haber, Nick
Krishna, Ranjay
Liu, Lingjie
Callison-Burch, Chris
Yatskar, Mark
Kembhavi, Aniruddha
Clark, Christopher
author_facet Yang, Yue
Sun, Fan-Yun
Weihs, Luca
VanderBilt, Eli
Herrasti, Alvaro
Han, Winson
Wu, Jiajun
Haber, Nick
Krishna, Ranjay
Liu, Lingjie
Callison-Burch, Chris
Yatskar, Mark
Kembhavi, Aniruddha
Clark, Christopher
contents 3D simulated environments play a critical role in Embodied AI, but their creation requires expertise and extensive manual effort, restricting their diversity and scope. To mitigate this limitation, we present Holodeck, a system that generates 3D environments to match a user-supplied prompt fully automatedly. Holodeck can generate diverse scenes, e.g., arcades, spas, and museums, adjust the designs for styles, and can capture the semantics of complex queries such as "apartment for a researcher with a cat" and "office of a professor who is a fan of Star Wars". Holodeck leverages a large language model (i.e., GPT-4) for common sense knowledge about what the scene might look like and uses a large collection of 3D assets from Objaverse to populate the scene with diverse objects. To address the challenge of positioning objects correctly, we prompt GPT-4 to generate spatial relational constraints between objects and then optimize the layout to satisfy those constraints. Our large-scale human evaluation shows that annotators prefer Holodeck over manually designed procedural baselines in residential scenes and that Holodeck can produce high-quality outputs for diverse scene types. We also demonstrate an exciting application of Holodeck in Embodied AI, training agents to navigate in novel scenes like music rooms and daycares without human-constructed data, which is a significant step forward in developing general-purpose embodied agents.
format Preprint
id arxiv_https___arxiv_org_abs_2312_09067
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Holodeck: Language Guided Generation of 3D Embodied AI Environments
Yang, Yue
Sun, Fan-Yun
Weihs, Luca
VanderBilt, Eli
Herrasti, Alvaro
Han, Winson
Wu, Jiajun
Haber, Nick
Krishna, Ranjay
Liu, Lingjie
Callison-Burch, Chris
Yatskar, Mark
Kembhavi, Aniruddha
Clark, Christopher
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Robotics
3D simulated environments play a critical role in Embodied AI, but their creation requires expertise and extensive manual effort, restricting their diversity and scope. To mitigate this limitation, we present Holodeck, a system that generates 3D environments to match a user-supplied prompt fully automatedly. Holodeck can generate diverse scenes, e.g., arcades, spas, and museums, adjust the designs for styles, and can capture the semantics of complex queries such as "apartment for a researcher with a cat" and "office of a professor who is a fan of Star Wars". Holodeck leverages a large language model (i.e., GPT-4) for common sense knowledge about what the scene might look like and uses a large collection of 3D assets from Objaverse to populate the scene with diverse objects. To address the challenge of positioning objects correctly, we prompt GPT-4 to generate spatial relational constraints between objects and then optimize the layout to satisfy those constraints. Our large-scale human evaluation shows that annotators prefer Holodeck over manually designed procedural baselines in residential scenes and that Holodeck can produce high-quality outputs for diverse scene types. We also demonstrate an exciting application of Holodeck in Embodied AI, training agents to navigate in novel scenes like music rooms and daycares without human-constructed data, which is a significant step forward in developing general-purpose embodied agents.
title Holodeck: Language Guided Generation of 3D Embodied AI Environments
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Robotics
url https://arxiv.org/abs/2312.09067