用视觉几何变压器提升大规模建图精度，显著减少轨迹漂移。
VGGT-SLAM++
- 基于VGGT的稠密平面数字高程图构建子地图并分块处理
- 通过DINOv2嵌入实现高效子地图融合，定位误差比之前低37%
- 适合需要高精度、低内存占用的自动驾驶与机器人场景
我们提出VGGT-SLAM++，一个完整的视觉SLAM系统，利用视觉几何接地变换器（VGGT）丰富的几何输出。系统包含融合VGGT前馈变换器与Sim(3)解的视觉里程计（前端）、基于数字高程图（DEM）的图构建模块，以及联合优化的大规模建图后端，实现高精度且内存受限的长期运行。相较于依赖稀疏回环闭合或全局Sim(3)流形约束的先前方法（如VGGT-SLAM），VGGT-SLAM++通过空间校正后端恢复高频局部束调整（LBA），显著降低短时漂移。对每个VGGT子地图，构建稠密平面-规范化的数字高程图（DEM），分割为块并计算DINOv2嵌入，以集成至共视图中。在共视窗口内使用视觉位置识别（VPR）模块检索空间邻居，触发频繁局部优化，稳定轨迹。在标准SLAM基准测试中，该系统达到最先进精度，显著减少短期漂移，加速图收敛，并以紧凑的DEM瓦片和次线性检索维持全局一致性。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。