发表机构
Jilin University; Beijing Institute of Control Engineering; Changchun University of Science and Technology; Shanghai Aerospace Control Technology Institute; Harbin Institute of Technology; The Hong Kong University of Science and Technology（吉林大学; 北京控制工程研究所; 长春理工大学; 上海航天控制技术研究所; 哈尔滨工业大学; 香港科技大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
MoonGS提出首个前馈3D高斯泼溅框架，利用图像对鲁棒深度特征实现月球表面高质量重建，在LuSNAR和MoonBlender数据集上超越现有方法，并保持亚秒级推理。
AI 中文摘要
从稀疏巡视器图像中进行高质量的月球地形三维重建对于自主月球探索至关重要，但由于视点重叠不足、表面纹理弱以及数据量有限，这仍然具有挑战性。我们提出MoonGS，这是首个专为月球场景定制的前馈三维高斯泼溅框架。仅给定两幅输入图像，MoonGS在单次前向传播中预测像素对齐的高斯原语，并在无需任何逐场景优化的情况下渲染出逼真的新视图。MoonGS（i）采用自适应骨干网络设计，无缝集成先进的视觉基础模型以提取鲁棒深度特征；（ii）以两种方式整合语义先验：将语义线索与视觉特征合并以细化高斯参数估计，并采用语义排序损失来正则化背景深度；（iii）采用熵引导的启发式重采样策略，通过选择信息量最大的远视点来增强稀疏观测，且开销可忽略不计。在LuSNAR基准和我们的合成弱纹理MoonBlender数据集上的实验表明，MoonGS在PSNR上超过最先进的前馈NeRF/3DGS基线+4.9 dB，SSIM+0.29，LPIPS降低40%，同时保持亚秒级推理。此外，我们通过展示框架有效利用包括VGGT在内的最先进骨干网络显著提升性能，验证了其广泛适用性。在嫦娥任务图像上的定性评估也显示，在比较方法中视觉质量最佳，表明其在真实月球数据上的鲁棒性。源代码和数据集在此https URL公开提供。
英文摘要
High-quality 3D reconstruction of lunar terrain from sparse rover images is indispensable for autonomous lunar exploration, but remains challenging because viewpoint overlap is insufficient, surface textures are weak, and data volume is limited. We propose MoonGS, the first feed-forward 3D Gaussian Splatting framework tailored to lunar scenes. Given only two input images, MoonGS predicts pixel-aligned Gaussian primitives in a single forward pass and renders photorealistic novel views without any per-scene optimization. MoonGS (i) adopts an adaptable backbone design that seamlessly integrates advanced vision foundation models to extract robust depth features; (ii) integrates semantic priors in two manners: merging semantic cues with visual features to refine Gaussian parameter estimation, and adopting a semantic ranking loss that regularizes background depth; and (iii) employs an entropy-guided heuristic resampling strategy to augment sparse observations by selecting the most informative distant viewpoints with negligible overhead. Experiments on the LuSNAR benchmark and our synthetic weak-texture MoonBlender dataset show that MoonGS surpasses state-of-the-art feed-forward NeRF/3DGS baselines by +4.9 dB PSNR, +0.29 SSIM, and 40\% lower LPIPS while maintaining sub-second inference. Furthermore, we validate the broad applicability of our framework by demonstrating that it effectively leverages state-of-the-art backbones, including VGGT, to significantly boost performance. Qualitative evaluations on Chang'e mission imagery also show the best visual quality among compared methods, indicating robustness on real lunar data. The source code and dataset are publicly available at https://github.com/InRobots/MoonBlender.
