Energy-GS：图像能量引导的位姿对齐高斯泼溅与重设计的位姿梯度流
Energy-GS: Image Energy-guided Pose Alignment Gaussian Splatting with redesigned pose gradient flow
针对 3DGS 基于光栅化的点渲染缺乏 NeRF 式可控采样、位姿梯度不稳定的问题，重新设计高斯图元的位姿梯度流并引入图像能量引导的渐进式位姿对齐约束，可直接用于无位姿先验的稀疏视图重建与 SLAM 中的位姿-地图联合优化。
原文依据 we redesign the optimization strategy of Gaussian primitives and introduce an image-energy-guided constraint that encourages progressive alignment of camera poses
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论辐射场中的高质量三维场景表示依赖于精确的相机位姿，而位姿在真实场景中往往难以获取。一种有效的解决方案是利用 RGB 图像对辐射场与相机位姿进行联合优化，这一思路在 NeRF 系列方法中已得到充分探索。然而与神经辐射场不同，三维高斯泼溅（3DGS）中的联合优化往往需要额外的正则化或先验空间知识才能达到可比性能。为消除这些依赖，我们提出 Energy-GS，一个仅使用 RGB 图像即可联合优化场景表示与相机位姿的位姿感知高斯泼溅框架。我们观察到，由于基于点的渲染机制，联合优化中的位姿梯度并不稳定。此外，不同于神经辐射场中可实现由粗到细位姿对齐的空间采样框架，基于光栅化的 3DGS 缺乏可控采样，因而无法支持渐进式位姿细化。为应对这些挑战，我们重新设计了高斯图元的优化策略，并引入图像能量引导的约束以促使相机位姿的渐进对齐。在合成与真实数据集上的实验表明，Energy-GS 能够同时有效优化场景重建并解决相机位姿失配问题。得益于仅依赖 RGB 图像，我们相信本工作为视觉定位与稠密建图等应用（如 SLAM）提供了有前景的启示。代码见 https://github.com/SkylerGao/ENGS 。
英文原文摘要 1625 字符
High-quality 3D scene representation in radiance fields relies on accurate camera poses which are often difficult to acquire in real-world scenarios. An effective solution is to use RGB images for the joint optimization of radiance fields and camera poses, an approach that has been well explored in NeRF series methods. However, unlike NeRF, joint optimization in 3D Gaussian Splatting (3DGS) often requires additional regularization or prior spatial knowledge to reach comparable performance. To eliminate these dependencies, we introduce Energy-GS, a pose-aware Gaussian splatting framework that jointly optimizes scene representation and camera poses using only RGB images. We observe that pose gradients in joint optimization are unstable due to the point-based rendering mechanism. Furthermore, unlike NeRF's spatial sampling framework that enables coarse-to-fine pose alignment, rasterization-based 3DGS lacks controllable sampling and thus cannot support progressive pose refinement. To address these challenges, we redesign the optimization strategy of Gaussian primitives and introduce an image-energy-guided constraint that encourages progressive alignment of camera poses. Experiments on both synthetic and real-world datasets show that Energy-GS can effectively optimize the scene reconstruction and resolve camera pose misalignment at the same time. Benefiting from reliance on only RGB images, we believe this work provides promising insights for visual localization and dense mapping applications such as SLAM. The Code is available at \href https://github.com/SkylerGao/ENGS https://github.com/SkylerGao/ENGS