arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2408.09042cs.CV
收藏

ADen:面向稀疏视图相机位姿估计的自适应密度表示

ADen: Adaptive Density Representations for Sparse-view Camera Pose Estimation

FAIR at Meta(Meta FAIR研究院)

Hao Tang, Weiyao Wang, Pierre Gleize, Matt Feiszli

AI总结:

针对稀疏视图下相机位姿估计中直接回归法不适定、概率法采样效率低的问题,提出结合生成器与判别器的ADen框架,兼具高精度与低运行时间优势。

摘要:

从一组图像中恢复相机位姿是3D计算机视觉的基础任务,为3D场景/物体重建等关键应用提供支撑。经典方法通常依赖特征对应关系(如关键点),要求输入图像具有大重叠度和小视角变化,这些要求在稀疏视图场景下带来了巨大挑战。 近年的数据驱动方法旨在直接输出相机位姿,要么通过回归6自由度(6DoF)相机位姿,要么将旋转建模为概率分布,但两类方法各有局限。一方面,直接回归相机位姿可能是不适定的,因为它假设分布为单模态,这在对称场景下不成立,会导致次优解。另一方面,概率方法能够建模对称歧义,但它们通过蛮力均匀采样整个旋转空间,这导致在提升模型精度的高采样密度与决定运行时间的采样效率之间存在不可避免的权衡。 本文提出ADen方法,通过引入生成器和判别器统一上述两种框架:生成器经训练输出6DoF相机位姿的多个假设,以表示分布并处理多模态歧义;判别器经训练识别出最能解释数据的假设。这使得ADen能够结合两类方法的优势,在实验评估中,其精度远高于此前方法,同时运行时间更短。

英文摘要:

Recovering camera poses from a set of images is a foundational task in 3D computer vision, which powers key applications such as 3D scene/object reconstructions. Classic methods often depend on feature correspondence, such as keypoints, which require the input images to have large overlap and small viewpoint changes. Such requirements present considerable challenges in scenarios with sparse views. Recent data-driven approaches aim to directly output camera poses, either through regressing the 6DoF camera poses or formulating rotation as a probability distribution. However, each approach has its limitations. On one hand, directly regressing the camera poses can be ill-posed, since it assumes a single mode, which is not true under symmetry and leads to sub-optimal solutions. On the other hand, probabilistic approaches are capable of modeling the symmetry ambiguity, yet they sample the entire space of rotation uniformly by brute-force. This leads to an inevitable trade-off between high sample density, which improves model precision, and sample efficiency that determines the runtime. In this paper, we propose ADen to unify the two frameworks by employing a generator and a discriminator: the generator is trained to output multiple hypotheses of 6DoF camera pose to represent a distribution and handle multi-mode ambiguity, and the discriminator is trained to identify the hypothesis that best explains the data. This allows ADen to combine the best of both worlds, achieving substantially higher precision as well as lower runtime than previous methods in empirical evaluations.

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。

↑