KITTI-360

Method

NeRF: Neural Radiance Fields [st] [NeRF]
https://github.com/bmild/nerf

Submitted on 6 Feb. 2022 16:05 by
Yiyi Liao (MPI)

Running time:10 s
Environment:1 core @ 2.5 Ghz (Python)

Method Description:
NeRF achieves state-of-the-art results for synthesizing
novel views of
complex scenes by optimizing an underlying continuous
volumetric
scene function using a sparse set of input views.
Parameters:
We train NeRF using the pytorch implementation:
https://github.com/yenchenlin/nerf-pytorch
Latex Bibtex:
@inproceedings{mildenhall2020nerf,
title={NeRF: Representing Scenes as Neural Radiance
Fields for View
Synthesis},
author={Ben Mildenhall and Pratul P. Srinivasan and
Matthew Tancik
and Jonathan T. Barron and Ravi Ramamoorthi and Ren
Ng},
year={2020},
booktitle={ECCV},
}

Detailed Results

This page provides detailed results for the method(s) selected. For the first 10 test images, we display the synthesized image and an error image. The error image visualizes the SSIM at every pixel. As the range of SSIM is within [-1,1] with 1 indicating the best performance and -1 indicating the worst, we visualize 1 - (1 + SSIM) / 2 such that bright region means large error and dark means low error.

Test Set Average

PSNR SSIM LPIPS
21.18 0.779 0.343
This table as LaTeX

Test Image 0

Prediction

Error


Test Image 1

Prediction

Error


Test Image 2

Prediction

Error


Test Image 3

Prediction

Error


Test Image 4

Prediction

Error


Test Image 5

Prediction

Error


Test Image 6

Prediction

Error


Test Image 7

Prediction

Error


Test Image 8

Prediction

Error


Test Image 9

Prediction

Error





eXTReMe Tracker