MELON：在SO(3)中利用未标定图像的NeRF
MELON: NeRF with Unposed Images in SO(3)
AI总结：
针对传统NeRF需精确相机位姿、未标定视角位姿估计非凸的问题，提出MELON方法，通过等价关系约简相机空间提升凸性，结合神经网络正则化，以更少视角实现高精度未标定图像NeRF重建。
摘要：
神经辐射场（Neural radiance fields，NeRF）能够从少量图像中实现具有照片级真实感的新视角合成和场景重建，但需要已知且精确的相机位姿。传统的位姿估计算法在平滑或自相似场景中会失效，而从未标定视角进行逆渲染的方法则需要相机朝向的粗略初始化。位姿估计的主要难点在于现实物体在某些变换下几乎具有不变性，这使得渲染视角之间的光度距离相对于相机参数是非凸的。我们利用一种与相机空间中局部极小值分布相匹配的等价关系，将该空间约简为其商集，在位姿估计中使其成为一个更具凸性的问题。通过使用神经网络对位姿估计进行正则化，我们证明了我们的方法——MELON——能够从未标定图像中重建神经辐射场，达到了当前最优的精度，同时所需视角数量仅为对抗式方法的十分之一。
英文摘要：
Neural radiance fields enable novel-view synthesis and scene reconstruction with photorealistic quality from a few images, but require known and accurate camera poses. Conventional pose estimation algorithms fail on smooth or self-similar scenes, while methods performing inverse rendering from unposed views require a rough initialization of the camera orientations. The main difficulty of pose estimation lies in real-life objects being almost invariant under certain transformations, making the photometric distance between rendered views non-convex with respect to the camera parameters. Using an equivalence relation that matches the distribution of local minima in camera space, we reduce this space to its quotient set, in which pose estimation becomes a more convex problem. Using a neural-network to regularize pose estimation, we demonstrate that our method - MELON - can reconstruct a neural radiance field from unposed images with state-of-the-art accuracy while requiring ten times fewer views than adversarial approaches.
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。
