EOGS++: Earth Observation Gaussian Splatting with Internal Camera Refinement and Direct Panchromatic Rendering

TL;DR

EOGS++ advances Earth observation 3D reconstruction by directly using high-res panchromatic satellite data, embedding bundle adjustment in training, and achieving state-of-the-art results efficiently.

Contribution

EOGS++ introduces a novel satellite imagery method with internal camera refinement and direct panchromatic rendering, improving accuracy and efficiency.

Findings

Achieves state-of-the-art reconstruction quality on IARPA 2016 and DFC2019 datasets.

Outperforms original EOGS and other NeRF-based methods in accuracy and speed.

Reduces mean MAE errors from 1.33 to 1.19 on buildings.

Abstract

Recently, 3D Gaussian Splatting has been introduced as a compelling alternative to NeRF for Earth observation, offering competitive reconstruction quality with significantly reduced training times. In this work, we extend the Earth Observation Gaussian Splatting (EOGS) framework to propose EOGS++, a novel method tailored for satellite imagery that directly operates on raw high-resolution panchromatic data without requiring external preprocessing. Furthermore, leveraging optical flow techniques we embed bundle adjustment directly within the training process, avoiding reliance on external optimization tools while improving camera pose estimation. We also introduce several improvements to the original implementation, including early stopping and TSDF post-processing, all contributing to sharper reconstructions and better geometric accuracy. Experiments on the IARPA 2016 and DFC2019 datasets demonstrate that EOGS++ achieves state-of-the-art performance in terms of reconstruction quality and efficiency, outperforming the original EOGS method and other NeRF-based methods while maintaining the computational advantages of Gaussian Splatting. Our model demonstrates an improvement from 1.33 to 1.19 mean MAE errors on buildings compared to the original EOGS models