Reconstruction pipeline measures lunar scene photorealistic accuracy
A Reproducible Reconstruction and Evaluation Framework for Photorealistic Lunar Analog Scene Reconstruction Toward Digital Twins
Robotics
Summary
Reconstructing detailed, realistic images of the Moon's surface is hard because rover cameras face challenges like motion blur and dust haze. This paper introduces a method to test how well 3D reconstructions from rover and human camera views match the real scenes. The authors found that common training measures don’t always predict final image quality when creating digital replicas of lunar terrain. They also created a new score to estimate reconstruction quality even without human reference data. Their work helps build better digital twins of the Moon for future exploration.
What this means in practice
- •For planetary rover engineers: Evaluate and improve 3D photorealistic mapping pipelines for lunar surface navigation and analysis by quantifying reconstruction accuracy under rover traversal effects.
- •For virtual reality developers: Enhance realistic digital twin environments of lunar analog scenes for immersive training or simulation by using the coverage score to assess view synthesis fidelity.
Authors
Antar Mazumder, Ryan Hartzell, Luke Boyd, Alexander Robinson, Neha Sharma, Chance Carnival, Frances Zhu
Abstract
In this paper, we present a reconstruction pipeline that empirically quantifies the photorealistic scene reconstruction accuracy from both human-controlled and rover mast camera monocular feeds against physical ground truth as a function of Lunar traversal effects (e.g., motion blur, haze due to lofted regolith, trajectory bias, and sparse views). First, we benchmark six variants of 3D Gaussian Splatting (3DGS) across three Lunar analog scenes on held-out test sets using standard photometric metrics. We then establish how under traversal effects, this test-set-specific Novel View Synthesis (NVS) performance does not proportionally scale to the same photometric metrics in the reconstructed DTs. Our results quantify that splat training metrics alone are insufficient predictors of Cross-Reconstruction Photorealistic NVS Fidelity (CRPNVS), and that factors such as rover trajectory bias and geometric scene coverage play a critical role in determining the photorealistic quality of the reconstructed scenes targeting Lunar analog Digital Twins. Additionally, we pilot study Coverage Score ($S_c$), a geometry-based diagnostic derived from COLMAP poses and point cloud structure that provides a meaningful heuristic about the highest achievable CRPNVS structural fidelity, independently of splat training performance in the absence of a human-controlled reference. These findings establish a standardized, human control-referenced evaluation pipeline for photorealistic scene reconstruction for Lunar-analog Digital Twin development and future planetary exploration. Code and project page: https://rose-lab-admin.github.io/rose-lunar-dt-rec