CUPID：联合物体与位姿建模的生成式三维重建
CUPID: Generative 3D Reconstruction via Joint Object and Pose Modeling
将相机位姿估计与物体生成联合建模为两阶段流模型，利用2D-3D对应关系鲁棒估计位姿，再以像素对齐特征条件化精化生成，为单图/多视图三维重建提供无需后优化的统一生成框架。
原文依据 Our two-stage flow-based model first generates a coarse 3D structure and 2D-3D correspondences to estimate the camera pose robustly.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出 Cupid，一种生成式三维重建框架，联合建模规范物体与相机位姿的完整分布。我们的两阶段流模型首先生成粗三维结构和2D-3D对应关系，以鲁棒地估计相机位姿。以该位姿为条件，精化阶段将像素对齐的图像特征直接注入生成过程，将生成模型的丰富先验与重建的几何保真度相结合。该策略实现了极高的保真度，在峰值信噪比上超过最优重建方法 3 dB 以上，在 Chamfer 距离上提升 10%。作为一个解耦物体与相机位姿的统一生成模型，Cupid 自然扩展到多视图和场景级重建任务，无需事后优化或微调。
英文原文摘要 848 字符
We introduce Cupid, a generative 3D reconstruction framework that jointly models the full distribution over both canonical objects and camera poses. Our two-stage flow-based model first generates a coarse 3D structure and 2D-3D correspondences to estimate the camera pose robustly. Conditioned on this pose, a refinement stage injects pixel-aligned image features directly into the generative process, marrying the rich prior of a generative model with the geometric fidelity of reconstruction. This strategy achieves exceptional faithfulness, outperforming state-of-the-art reconstruction methods by over 3 dB PSNR and 10% in Chamfer Distance. As a unified generative model that decouples the object and camera pose, Cupid naturally extends to multi-view and scene-level reconstruction tasks without requiring post-hoc optimization or fine-tuning.