arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2402.08437cs.CV
收藏

通过旋转矩阵与投影矩阵的几何约束进行相机标定

Camera Calibration through Geometric Constraints from Rotation and Projection Matrices

Graphics Lab, LUMS(拉合尔管理科学大学图形实验室)

Muhammad Waleed, Abdul Rauf, Murtaza Taj

AI总结:

本文提出一种基于投影矩阵几何性质和旋转矩阵正交性的无监督几何约束损失(UGCL),通过多任务学习混合估计相机内外参,并构建含900多种配置、63,600对图像的CVGL数据集,在合成与真实数据上均优于SOTA方法。

摘要:

相机标定过程涉及估计内参和外参,这些参数对于准确执行三维重建、目标跟踪和增强现实等任务至关重要。在本工作中,我们提出了一种新颖的基于约束的损失函数,用于度量内参(焦距:$(f_x, f_y)$ 和主点:$(p_x, p_y)$)以及外参(基线:($b$)、视差:($d$)、平移:$(t_x, t_y, t_z)$,以及旋转,特别是俯仰角:$(θ_p)$)相机参数。我们的新颖约束基于相机模型固有的几何性质,包括投影矩阵的结构(消失点、世界原点图像、轴平面)以及旋转矩阵的正交性。因此,我们通过多任务学习框架提出了一种新颖的无监督几何约束损失(UGCL)。我们的方法是一种混合方法,利用神经网络的学习能力来估计所需参数,同时结合相机投影矩阵中固有的底层数学性质。这种独特的方法不仅增强了模型的可解释性,而且促进了更具信息性的学习过程。此外,我们引入了一个新的 CVGL 相机标定数据集,包含超过 900 种相机参数配置,并纳入 63,600 对紧密反映真实世界条件的图像对。通过在合成数据集和真实世界数据集上进行训练和测试,我们提出的方法与最先进(SOTA)基准相比,在所有参数上都表现出改进。代码和更新后的数据集可在此处找到:https://github.com/CVLABLUMS/CVGL-Camera-Calibration

英文摘要:

The process of camera calibration involves estimating the intrinsic and extrinsic parameters, which are essential for accurately performing tasks such as 3D reconstruction, object tracking and augmented reality. In this work, we propose a novel constraints-based loss for measuring the intrinsic (focal length: $(f_x, f_y)$ and principal point: $(p_x, p_y)$) and extrinsic (baseline: ($b$), disparity: ($d$), translation: $(t_x, t_y, t_z)$, and rotation specifically pitch: $(θ_p)$) camera parameters. Our novel constraints are based on geometric properties inherent in the camera model, including the anatomy of the projection matrix (vanishing points, image of world origin, axis planes) and the orthonormality of the rotation matrix. Thus we proposed a novel Unsupervised Geometric Constraint Loss (UGCL) via a multitask learning framework. Our methodology is a hybrid approach that employs the learning power of a neural network to estimate the desired parameters along with the underlying mathematical properties inherent in the camera projection matrix. This distinctive approach not only enhances the interpretability of the model but also facilitates a more informed learning process. Additionally, we introduce a new CVGL Camera Calibration dataset, featuring over 900 configurations of camera parameters, incorporating 63,600 image pairs that closely mirror real-world conditions. By training and testing on both synthetic and real-world datasets, our proposed approach demonstrates improvements across all parameters when compared to the state-of-the-art (SOTA) benchmarks. The code and the updated dataset can be found here: https://github.com/CVLABLUMS/CVGL-Camera-Calibration

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。

↑