ConfidentSplat：面向精确3D高斯泼溅SLAM的置信度加权深度融合
ConfidentSplat: Confidence-Weighted Depth Fusion for Accurate 3D Gaussian Splatting SLAM
AI总结：
ConfidentSplat提出置信度加权深度融合机制，结合多视图几何与单目先验生成高保真代理深度，指导可变形3DGS地图优化，在纯RGB稠密SLAM中显著提升重建精度与新视角合成质量。
摘要：
我们提出ConfidentSplat，一种新颖的基于3D高斯泼溅（3DGS）的SLAM系统，用于鲁棒、高保真的纯RGB重建。针对现有纯RGB 3DGS SLAM方法中因不可靠深度估计而产生的几何不准确问题，ConfidentSplat引入了一项核心创新：置信度加权融合机制。该机制自适应地融合来自多视图几何的深度线索与学习的单目先验（Omnidata ViT），基于显式的可靠性估计——主要来源于多视图几何一致性——动态加权其贡献，以生成用于地图监督的高保真代理深度。由此产生的代理深度指导可变形3DGS地图的优化，该地图在受DROID-SLAM启发的后端优化（闭环检测、全局光束法平差）和前端位姿更新后，能够在线高效适应以保持全局一致性。在标准基准（TUM-RGBD、ScanNet）和多样的自采集移动数据集上的广泛验证表明，与基线方法相比，在重建精度（L1深度误差）和新视角合成保真度（PSNR、SSIM、LPIPS）方面均有显著提升，尤其在具有挑战性的条件下。ConfidentSplat凸显了有原则的、置信度感知的传感器融合对于推进最先进稠密视觉SLAM的有效性。
英文摘要：
We introduce ConfidentSplat, a novel 3D Gaussian Splatting (3DGS)-based SLAM system for robust, highfidelity RGB-only reconstruction. Addressing geometric inaccuracies in existing RGB-only 3DGS SLAM methods that stem from unreliable depth estimation, ConfidentSplat incorporates a core innovation: a confidence-weighted fusion mechanism. This mechanism adaptively integrates depth cues from multiview geometry with learned monocular priors (Omnidata ViT), dynamically weighting their contributions based on explicit reliability estimates-derived predominantly from multi-view geometric consistency-to generate high-fidelity proxy depth for map supervision. The resulting proxy depth guides the optimization of a deformable 3DGS map, which efficiently adapts online to maintain global consistency following pose updates from a DROID-SLAM-inspired frontend and backend optimizations (loop closure, global bundle adjustment). Extensive validation on standard benchmarks (TUM-RGBD, ScanNet) and diverse custom mobile datasets demonstrates significant improvements in reconstruction accuracy (L1 depth error) and novel view synthesis fidelity (PSNR, SSIM, LPIPS) over baselines, particularly in challenging conditions. ConfidentSplat underscores the efficacy of principled, confidence-aware sensor fusion for advancing state-of-the-art dense visual SLAM.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
