PreF3R：从变长图像序列中实现无姿态前馈3D高斯溅射
PreF3R: Pose-Free Feed-Forward 3D Gaussian Splatting from Variable-length Image Sequence
AI总结：
针对无姿态变长图像序列的3D重建问题，提出PreF3R方法，无需相机校准，通过空间记忆网络扩展DUSt3R至多视图输入，结合密集高斯参数预测头，实现20 FPS实时重建与高效新视角渲染，泛化性强。
摘要：
我们提出PreF3R，即基于变长图像序列的无姿态前馈3D重建方法。与现有方法不同，PreF3R无需相机校准，直接从无姿态图像序列在规范坐标系内重建3D高斯场，实现高效新视角渲染。我们利用DUSt3R的成对3D结构重建能力，通过空间记忆网络将其扩展到连续多视图输入，无需基于优化的全局对齐。此外，PreF3R集成密集高斯参数预测头，支持后续通过可微分光栅化进行新视角合成，结合光度损失和点图回归损失监督模型，提升真实感和结构精度。给定有序图像序列，PreF3R以20 FPS增量重建3D高斯场，实现实时新视角渲染。实验表明，PreF3R能有效解决无姿态前馈新视角合成难题，且对未见过的场景泛化性强。
英文摘要：
We present PreF3R, Pose-Free Feed-forward 3D Reconstruction from an image sequence of variable length. Unlike previous approaches, PreF3R removes the need for camera calibration and reconstructs the 3D Gaussian field within a canonical coordinate frame directly from a sequence of unposed images, enabling efficient novel-view rendering. We leverage DUSt3R's ability for pair-wise 3D structure reconstruction, and extend it to sequential multi-view input via a spatial memory network, eliminating the need for optimization-based global alignment. Additionally, PreF3R incorporates a dense Gaussian parameter prediction head, which enables subsequent novel-view synthesis with differentiable rasterization. This allows supervising our model with the combination of photometric loss and pointmap regression loss, enhancing both photorealism and structural accuracy. Given a sequence of ordered images, PreF3R incrementally reconstructs the 3D Gaussian field at 20 FPS, therefore enabling real-time novel-view rendering. Empirical experiments demonstrate that PreF3R is an effective solution for the challenging task of pose-free feed-forward novel-view synthesis, while also exhibiting robust generalization to unseen scenes.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
