SparseSat-NeRF: Dense Depth Supervised Neural Radiance Fields for Sparse Satellite Images

TL;DR

SparseSat-NeRF introduces a novel method that enhances neural radiance fields for satellite images with sparse views by incorporating dense depth supervision from traditional stereo matching, improving 3D reconstruction accuracy.

Contribution

The paper presents SparseSat-NeRF, an extension of Sat-NeRF that effectively utilizes sparse satellite views with dense depth supervision, addressing limitations of existing NeRF methods in earth observation.

Findings

Outperforms NeRF and Sat-NeRF on satellite imagery

Effective with sparse stereo and tri-stereo images

Improves scene geometry reconstruction accuracy

Abstract

Digital surface model generation using traditional multi-view stereo matching (MVS) performs poorly over non-Lambertian surfaces, with asynchronous acquisitions, or at discontinuities. Neural radiance fields (NeRF) offer a new paradigm for reconstructing surface geometries using continuous volumetric representation. NeRF is self-supervised, does not require ground truth geometry for training, and provides an elegant way to include in its representation physical parameters about the scene, thus potentially remedying the challenging scenarios where MVS fails. However, NeRF and its variants require many views to produce convincing scene's geometries which in earth observation satellite imaging is rare. In this paper we present SparseSat-NeRF (SpS-NeRF) - an extension of Sat-NeRF adapted to sparse satellite views. SpS-NeRF employs dense depth supervision guided by crosscorrelation similarity metric provided by traditional semi-global MVS matching. We demonstrate the effectiveness of our approach on stereo and tri-stereo Pleiades 1B/WorldView-3 images, and compare against NeRF and Sat-NeRF. The code is available at https://github.com/LulinZhang/SpS-NeRF