期刊
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
卷 -, 期 -, 页码 8238-8248出版社
IEEE COMPUTER SOC
DOI: 10.1109/CVPR52688.2022.00807
关键词
-
We propose a variant called Block-NeRF to represent large-scale environments. By decomposing the scene into individually trained NeRFs, rendering time can be decoupled from scene size, enabling rendering in arbitrarily large environments. We also introduce several architectural changes to make NeRF robust to different environmental conditions, and a procedure for aligning appearance between adjacent NeRFs for seamless combination.
We present Block-NeRF, a variant of Neural Radiance Fields that can represent large-scale environments. Specifically, we demonstrate that when scaling NeRF to render city-scale scenes spanning multiple blocks, it is vital to decompose the scene into individually trained NeRFs. This decomposition decouples rendering time from scene size, enables rendering to scale to arbitrarily large environments, and allows per-block updates of the environment. We adopt several architectural changes to make NeRF robust to data captured over months under different environmental conditions. We add appearance embeddings, learned pose refinement, and controllable exposure to each individual NeRF, and introduce a procedure for aligning appearance between adjacent NeRFs so that they can be seamlessly combined. We build a grid of Block-NeRFs from 2.8 million images to create the largest neural scene representation to date, capable of rendering an entire neighborhood of San Francisco.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据