提出轻量级实时位姿估计方法，解决旋转与平移耦合问题。
IUP-Pose: Decoupled Iterative Uncertainty Propagation for Real-time Relative Pose Regression via Implicit Dense Alignment v1
- 分步解耦旋转与平移估计，迭代优化并传播不确定性
- 在MegaDepth1500上达73.3% AUC@20deg，推理速度70 FPS
- 适合边缘设备实时部署，仅37M参数
相对位姿估计是SLAM、视觉定位和三维重建的基础。现有相对位姿回归（RPR）方法面临核心权衡：基于特征匹配的流程精度高但因不可微的RANSAC阻断梯度流；而基于ViT的回归器虽可端到端训练，却难以实现实时部署。我们识别出关键瓶颈在于旋转与平移估计的耦合及跨视图特征对齐不足。提出IUP-Pose，一种基于几何先验的解耦迭代框架，含隐式密集对齐机制。轻量级多头双向交叉注意力（MHBC）模块在无显式匹配监督下对齐跨视图特征。对齐后的特征通过解耦的旋转-平移流水线处理：两个共享参数的旋转阶段迭代细化旋转并传播不确定性，旋转同态变换H_inf用于重对齐特征图后预测平移。IUP-Pose在MegaDepth1500上实现73.3% AUC@20deg，吞吐率达70 FPS，仅需37M参数，展现出优异的精度-效率权衡，适用于实时边缘部署。
原文摘要 · Abstract (English)
Relative pose estimation is fundamental for SLAM, visual localization, and 3D reconstruction. Existing Relative Pose Regression (RPR) methods face a key trade-off: feature-matching pipelines achieve high accuracy but block gradient flow via non-differentiable RANSAC, while ViT-based regressors are end-to-end trainable but prohibitively expensive for real-time deployment. We identify the core bottlenecks as the coupling between rotation and translation estimation and insufficient cross-view feature alignment. We propose IUP-Pose, a geometry-driven decoupled iterative framework with implicit dense alignment. A lightweight Multi-Head Bi-Cross Attention (MHBC) module aligns cross-view features without explicit matching supervision. The aligned features are processed by a decoupled rotation-translation pipeline: two shared-parameter rotation stages iteratively refine rotation with uncertainty, and feature maps are realigned via rotational homography H_inf before translation prediction. IUP-Pose achieves 73.3% AUC@20deg on MegaDepth1500 with full end-to-end differentiability, 70 FPS throughput, and only 37M parameters, demonstrating a favorable accuracy-efficiency trade-off for real-time edge deployment.
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。