VGGT-SLAM++：基于视觉几何基础变换器的视觉SLAM系统
VGGT-SLAM++
AI总结：
VGGT-SLAM++结合视觉几何基础变换器的几何丰富输出，通过空间校正后端实现高频率局部捆绑调整，提升大规模映射精度与内存效率。
摘要：
我们介绍了VGGT-SLAM++，一种利用视觉几何基础变换器（VGGT）几何丰富输出的完整视觉SLAM系统。该系统包括融合VGGT前馈变换器和Sim(3)求解器的视觉里程计（前端），基于数字高程图（DEM）的图构建模块，以及联合实现精确大规模映射与受限内存的后端。与以往基于变换器的SLAM流程如VGGT-SLAM主要依赖稀疏回环闭合或全局Sim(3)流形约束导致短时间姿态漂移不同，VGGT-SLAM++通过空间校正后端恢复高频率局部捆绑调整（LBA）。对于每个VGGT子地图，我们构建密集平面-标准DEM，将其划分为补丁并计算其DINOv2嵌入以将子地图整合到协可视图中。通过协可视窗口内的视觉位置识别（VPR）模块检索空间邻居，触发频繁局部优化以稳定轨迹。在标准SLAM基准测试中，VGGT-SLAM++实现了最先进的准确性，显著减少了短期漂移，加速了图收敛，并通过紧凑的DEM瓷砖和子线性检索保持了全局一致性。
英文摘要：
We introduce VGGT-SLAM++, a complete visual SLAM system that leverages the geometry-rich outputs of the Visual Geometry Grounded Transformer (VGGT). The system comprises a visual odometry (front-end) fusing the VGGT feed-forward transformer and a Sim(3) solution, a Digital Elevation Map (DEM)-based graph construction module, and a back-end that jointly enable accurate large-scale mapping with bounded memory. While prior transformer-based SLAM pipelines such as VGGT-SLAM rely primarily on sparse loop closures or global Sim(3) manifold constraints - allowing short-horizon pose drift - VGGT-SLAM++ restores high-cadence local bundle adjustment (LBA) through a spatially corrective back-end. For each VGGT submap, we construct a dense planar-canonical DEM, partition it into patches, and compute their DINOv2 embeddings to integrate the submap into a covisibility graph. Spatial neighbors are retrieved using a Visual Place Recognition (VPR) module within the covisibility window, triggering frequent local optimization that stabilizes trajectories. Across standard SLAM benchmarks, VGGT-SLAM++ achieves state-of-the-art accuracy, substantially reducing short-term drift, accelerating graph convergence, and maintaining global consistency with compact DEM tiles and sublinear retrieval.
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。
