外观
论文arXiv 2609.214371 个月前
局部思考，全局优化以实现内存高效的3D重建
原文链接：Think Locally, Refine Globally for Memory-Efficient 3D Reconstruction（arXiv 2609.21437 · 发布日期：2026-09-18）
Think Locally, Refine Globally for Memory-Efficient 3D Reconstruction
LoG-VGGT面向长序列3D重建，结合跨窗口注意力与全局相机一致性优化，在多个长序列基准上验证深度与位姿表现。
核心要点
- 长序列3D重建需兼顾局部时序建模、全局相机一致性与受限内存。
- 在少量Transformer块引入跨窗口注意力，使相邻时间窗口传播信息并限制内存。
- 通过相机token与紧凑register token交叉注意力，缓解长期位姿漂移。
- 大量实验显示其在多个长序列基准上提升深度精度与相机位姿估计稳健性。
验证范围与局限：验证范围为多个长序列基准上的实验和流式重建性能评估。
英文摘要 (Abstract)
We propose LoG-VGGT, a memory-efficient framework for long-sequence 3D reconstruction that balances local temporal modeling with global camera consistency. Instead of relying on full global attention, our method introduces cross-window attention at a small subset of transformer blocks, enabling effective information propagation across adjacent temporal windows while keeping memory usage bounded. To mitigate long-term pose drift, we further design a global camera consistency refinement module, where camera tokens interact with compact register tokens via cross-attention to enforce scene-level constraints across the entire sequence. This design enables joint optimization of camera representations and significantly improves long-horizon pose stability without incurring the high cost of sequence-wide attention. Extensive experiments demonstrate that LoG-VGGT achieves improved depth accuracy and robust camera pose estimation across multiple long-sequence benchmarks, while delivering competitive streaming reconstruction performance.
作者：Jingke Zhou, Chenhang Ma, Zhizhou Zhong, Mingkai Liu, Zhuang Zhou, Yicheng Ji, Binghua Su, Bo Cai, Xianliang Huang
来源
本页事实对应的公开页面，以原文为准。