[2022] Block-NeRF: Scalable Large Scene Neural View Synthesis

[gatheluck]

論文リンク

• arXiv
• project page

公開日（yyyy/mm/dd）

2022/02/10

概要

TeX

% yyyy/mm/dd
@article{
    tancik2022block,
    title={Block-NeRF: Scalable Large Scene Neural View Synthesis},
    author={Matthew Tancik and Vincent Casser and Xinchen Yan and Sabeek Pradhan and Ben Mildenhall and Pratul P. Srinivasan and Jonathan T. Barron and Henrik Kretzschmar},
    journal=arXiv # "2202.05263",
    year={2022}
}

[gatheluck]

Abstract

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.

Anatomy

• Explain what they propose in short
• Describe key idea (Issue statement)
• Explain merit of key idea
• Describe additional new points
• Explain detail and merit of new points

Type

• New method + Impressive result