DAGE：面向高效细粒度几何估计的双流架构
DAGE: Dual-Stream Architecture for Efficient and Fine-Grained Geometry Estimation
用低分辨率全局流负责跨视图一致性与相机估计、高分辨率逐帧流保留细节，再以轻量适配器跨注意力融合，为高分辨率长序列的深度/点图与位姿联合估计提供了可复用的解耦式双流范式。
原文依据 A lightweight adapter fuses these streams via cross-attention, injecting global context without disturbing the pretrained single-frame pathway.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论从无标定的多视图/视频输入中估计准确且视角一致的几何与相机位姿仍然具有挑战性——尤其在高空间分辨率和长序列情形下。我们提出 DAGE，一种双流 Transformer，其主要创新在于将全局一致性与精细细节解耦。低分辨率流在经大幅下采样的帧上运行，采用交替的帧级/全局注意力来构建视角一致的表示并高效估计相机；高分辨率流则逐帧处理原始图像，以保留锐利边界与小结构。一个轻量适配器通过跨注意力融合这两条流，在不干扰预训练单帧通路的前提下注入全局上下文。该设计可独立扩展分辨率与片段长度，支持最高 2K 的输入，并保持实用的推理开销。DAGE 输出锐利的深度/点图、强跨视图一致性以及准确的位姿，在视频几何估计与多视图重建上取得了新的最优结果。
英文原文摘要 1072 字符
Estimating accurate, view-consistent geometry and camera poses from uncalibrated multi-view/video inputs remains challenging--especially at high spatial resolutions and over long sequences. We present DAGE, a dual-stream transformer whose main novelty is to disentangle global coherence from fine detail. A low-resolution stream operates on aggressively downsampled frames with alternating frame/global attention to build a view-consistent representation and estimate cameras efficiently, while a high-resolution stream processes the original images per-frame to preserve sharp boundaries and small structures. A lightweight adapter fuses these streams via cross-attention, injecting global context without disturbing the pretrained single-frame pathway. This design scales resolution and clip length independently, supports inputs up to 2K, and maintains practical inference cost. DAGE delivers sharp depth/pointmaps, strong cross-view consistency, and accurate poses, establishing new state-of-the-art results for video geometry estimation and multi-view reconstruction.