跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.26493 · v1

SGAR: Feed-Forward 3D Reconstruction via Sparse Geometry-Guided Attention

SGAR:基于稀疏几何引导注意力的前馈三维重建

首次提交 2026-11-20 01:46
版本更新 2026-11-20 01:46
雷达观测 2026-11-20 12:52
arXiv ↗PDF ↗

摘要

Feed-forward models have been very successful in 3D reconstruction with ad hoc attention designs that relate all view patches densely. However, the cost of this dense global attention grows quadratically with the number of views, preventing the model from applying to larger-scale reconstruction settings. In this paper, we argue that most dense attention computations are unnecessary from a classical structure-from-motion perspective: 3D structures stem from geometrically matching regions in 2D. We bring this principle to learning-based feed-forward reconstruction with our Sparse Geometric Attention Reconstruction (SGAR) method. In SGAR, each patch token attends to only a small set of patch tokens from other views, chosen by a lightweight learned key selection module trained to predict cross-view 3D proximity of tokens. Thus, dense global attention is replaced by much more efficient geometric-similarity-guided sparse attention. SGAR is a general method that can be applied to different feed-forward models by fine-tuning. In our experiments, we show that SGAR is the fastest among both state-of-the-art dense and sparse methods and the most accurate sparse one. It runs up to 12-times faster than dense attention at 1,000 views with stable pose accuracy where existing sparse methods fail, and shows the best performance in reconstructing more complete geometries on the same time budget.

AI 论文解读

基于摘要 · 非全文审阅

SGAR 用轻量学习式关键选择模块预测跨视角三维邻近性,把稠密全局注意力替换为几何相似性引导的稀疏注意力,从而让前馈三维重建在更多视角下更快且更准。

正在读取账号与研究记录…

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-20 01:46 · 最新观测版本

SGAR: Feed-Forward 3D Reconstruction via Sparse Geometry-Guided Attention

Feed-forward models have been very successful in 3D reconstruction with ad hoc attention designs that relate all view patches densely. However, the cost of this dense global attention grows quadratically with the number of views, preventing the model from applying to larger-scale reconstruction settings. In this paper, we argue that most dense attention computations are unnecessary from a classical structure-from-motion perspective: 3D structures stem from geometrically matching regions in 2D. We bring this principle to learning-based feed-forward reconstruction with our Sparse Geometric Attention Reconstruction (SGAR) method. In SGAR, each patch token attends to only a small set of patch tokens from other views, chosen by a lightweight learned key selection module trained to predict cross-view 3D proximity of tokens. Thus, dense global attention is replaced by much more efficient geometric-similarity-guided sparse attention. SGAR is a general method that can be applied to different feed-forward models by fine-tuning. In our experiments, we show that SGAR is the fastest among both state-of-the-art dense and sparse methods and the most accurate sparse one. It runs up to 12-times faster than dense attention at 1,000 views with stable pose accuracy where existing sparse methods fail, and shows the best performance in reconstructing more complete geometries on the same time budget.

查看此版本解读 →查看此版本原文 ↗