CaesarNeRF: Calibrated Semantic Representation for Few-shot Generalizable Neural Rendering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Haidong, Ding, Tianyu, Chen, Tianyi, Zharkov, Ilya, Nevatia, Ram, Liang, Luming
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909249063354368
author Zhu, Haidong
Ding, Tianyu
Chen, Tianyi
Zharkov, Ilya
Nevatia, Ram
Liang, Luming
author_facet Zhu, Haidong
Ding, Tianyu
Chen, Tianyi
Zharkov, Ilya
Nevatia, Ram
Liang, Luming
contents Generalizability and few-shot learning are key challenges in Neural Radiance Fields (NeRF), often due to the lack of a holistic understanding in pixel-level rendering. We introduce CaesarNeRF, an end-to-end approach that leverages scene-level CAlibratEd SemAntic Representation along with pixel-level representations to advance few-shot, generalizable neural rendering, facilitating a holistic understanding without compromising high-quality details. CaesarNeRF explicitly models pose differences of reference views to combine scene-level semantic representations, providing a calibrated holistic understanding. This calibration process aligns various viewpoints with precise location and is further enhanced by sequential refinement to capture varying details. Extensive experiments on public datasets, including LLFF, Shiny, mip-NeRF 360, and MVImgNet, show that CaesarNeRF delivers state-of-the-art performance across varying numbers of reference views, proving effective even with a single reference image.
format Preprint
id arxiv_https___arxiv_org_abs_2311_15510
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle CaesarNeRF: Calibrated Semantic Representation for Few-shot Generalizable Neural Rendering
Zhu, Haidong
Ding, Tianyu
Chen, Tianyi
Zharkov, Ilya
Nevatia, Ram
Liang, Luming
Computer Vision and Pattern Recognition
Generalizability and few-shot learning are key challenges in Neural Radiance Fields (NeRF), often due to the lack of a holistic understanding in pixel-level rendering. We introduce CaesarNeRF, an end-to-end approach that leverages scene-level CAlibratEd SemAntic Representation along with pixel-level representations to advance few-shot, generalizable neural rendering, facilitating a holistic understanding without compromising high-quality details. CaesarNeRF explicitly models pose differences of reference views to combine scene-level semantic representations, providing a calibrated holistic understanding. This calibration process aligns various viewpoints with precise location and is further enhanced by sequential refinement to capture varying details. Extensive experiments on public datasets, including LLFF, Shiny, mip-NeRF 360, and MVImgNet, show that CaesarNeRF delivers state-of-the-art performance across varying numbers of reference views, proving effective even with a single reference image.
title CaesarNeRF: Calibrated Semantic Representation for Few-shot Generalizable Neural Rendering
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.15510