arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.16936cs.CVcs.LG
收藏

基于布局学习的解耦三维场景生成

Disentangled 3D Scene Generation with Layout Learning

UC Berkeley(加州大学伯克利分校);Google Research(谷歌研究院)

Dave Epstein, Ben Poole, Ben Mildenhall, Alexei A. Efros, Aleksander Holynski

AI总结:

本文提出一种无监督解耦三维场景生成方法,通过联合优化多个NeRF和布局,并利用预训练文本到图像模型引导,将场景分解为独立对象,从而增强文本到三维内容创建能力。

摘要:

我们提出了一种生成三维场景的方法,该场景被解耦为其组成对象。这种解耦是无监督的,仅依赖于大型预训练文本到图像模型的知识。我们的核心见解是,可以通过寻找三维场景中的某些部分来发现对象,这些部分在空间上重新排列后,仍然能产生同一场景的有效配置。具体而言,我们的方法从头开始联合优化多个NeRF——每个NeRF代表其自身的对象——以及一组将这些对象合成为场景的布局。然后,我们鼓励这些合成场景根据图像生成器处于分布内。我们表明,尽管方法简单,我们的方法成功地生成了分解为单个对象的三维场景,从而在文本到三维内容创建中实现了新的能力。有关结果和交互式演示,请参见我们的项目页面:https://dave.ml/layoutlearning/

英文摘要:

We introduce a method to generate 3D scenes that are disentangled into their component objects. This disentanglement is unsupervised, relying only on the knowledge of a large pretrained text-to-image model. Our key insight is that objects can be discovered by finding parts of a 3D scene that, when rearranged spatially, still produce valid configurations of the same scene. Concretely, our method jointly optimizes multiple NeRFs from scratch - each representing its own object - along with a set of layouts that composite these objects into scenes. We then encourage these composited scenes to be in-distribution according to the image generator. We show that despite its simplicity, our approach successfully generates 3D scenes decomposed into individual objects, enabling new capabilities in text-to-3D content creation. For results and an interactive demo, see our project page at https://dave.ml/layoutlearning/

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑