SceneDreamer: unbounded 3D scene generation from 2D image collections

• Main Authors: Chen, Zhaoxi, Wang, Guangcong, Liu, Ziwei
• Other Authors: School of Computer Science and Engineering
• Format: Article
• Language: English
• Published: 2024
• Subjects: Computer and Information Science; Neural Rendering; Unbounded Scene Generation
• Online Access: https://hdl.handle.net/10356/173443
• Institution: Nanyang Technological University

In this work, we present SceneDreamer, an unconditional generative model for unbounded 3D scenes, which synthesizes large-scale 3D landscapes from random noise. Our framework is learned from in-the-wild 2D image collections only, without any 3D annotations. At the core of SceneDreamer is a principle...