DAGE: Dual-Stream Architecture for Efficient and Fine-Grained Geometry Estimation
Ngo, Tuan Duc, Huang, Jiahui, Oh, Seoung Wug, Blackburn-Matzen, Kevin, Kalogerakis, Evangelos, Gan, Chuang, Lee, Joon-Young
原文摘要
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.
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表