基于无监督级联的亚像素单应矩阵估计

为提高单应性估计的准确性和解决真实标注难获取的问题, 提出一种具有修正功能的无监督单应性估计算法。该算法采用级联结构, 其思想类似于迭代, 其中每一级网络都保持相同的层数和参数量, 下一级网络输出的单应性矩阵为真实矩阵与之前输出单应矩阵和的残差。考虑到模型复杂度和实时性的需求, 文章采用两级网络级联。通过在COCO数据集中的5000张图片上进行验证, 结果表明, 相比传统方法和其他基于深度学习的方法, 所设计的级联无监督算法具有更准确的估计能力, 其在测试集中的平均像素误差为0.54, 较传统方法下降95.38%, 运行速度达到95f/s。

Abstract

In order to improve the accuracy of homography estimation and solve the problem that it is difficult to obtain real labels, an unsupervised homography estimation algorithm with correction function is proposed. The algorithm adopts a cascade structure, and its idea is similar to iteration, in which each level network maintains the same number of layers and parameters, and the homography of the output of the next level network is estimated as the residual of the sum of the real matrix and the previous output homography matrix. Considering the requirements of model complexity and real-time, a two-level network cascade is adopted in this paper. Through the verification on 5000 images in the COCO dataset, it shows that the cascade unsupervised algorithm designed in this paper has more accurate estimation ability than traditional methods and other methods based on deep learning. Its average pixel error in the test set is 0.54, which is 95.38% lower than that of traditional methods, and the running speed reaches 95f/s.

参考文献

[1] Zhao Q, Ma Y, Zhu C, et al. Image stitching via deep homography estimation［J］. Neurocomputing, 2021, 450: 219-229.

[2] Mur-Artal R, Montiel J M M, Tardos J D. ORB-SLAM: A versatile and accurate monocular SLAM system［J］. IEEE Trans. on Robotics, 2015, 31(5): 1147-1163.

[3] 鲁甜, 刘蓉, 刘明, 等. 基于特征图注意力机制的图像超分辨率重建［J］. 计算机工程, 2021, 47(3): 261-268.

[4] Wu R, Xu Z, Zhang J, et al. Robust global motion estimation for video stabilization based on improved K-means clustering and superpixel［J］. Sensors, 2021, 21(7): 2505.

[5] 张墨华, 彭建华. 面向图像先验建模的可扩展高斯混合模型［J］. 计算机工程, 2020, 46(4): 220-227.

[6] Lowe D G. Distinctive image features from scale-invariant keypoints［J］. Inter. J. of Computer Vision, 2004, 60(2): 91-110.

[7] Rublee E, Rabaud V, Konolige K, et al. ORB: an efficient alternative to SIFT or SURF［C］// IEEE Inter. Conf. on Computer Vision, 2011.

[8] Szeliski R. Image alignment and stitching: A tutorial［J］. Foundations and Trends in Computer Graphics and Vision, 2006, 2(1): 1-104.

[9] Fischler M A, Bolles R C. Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography-ScienceDirect［J］. Readings in Computer Vision, 1987: 726-740.

[10] Detone D, Malisiewicz T, Rabinovich A. Deep image homography estimation［J］. RSS Workshop on Limits and Potentials of Deep Learning in Robotics, 2016: arXiv: 1606.03798.

[11] Japkowicz N, Nowruzi F E, Laganiere R. Homography estimation from image pairs with hierarchical convolutional networks［C］// The IEEE Inter. Conf. on Computer Vision (ICCV), 2017.

[12] Nguyen T, Chen S W, Shivakumar S S, et al. Unsupervised deep homography: A fast and robust homography estimation model［J］. IEEE Robotics and Automation Letters, 2017: arXiv: 1709.03966.

[13] 郭孟夏. 一种基于深度学习的视频单应性估计算法［J］. 现代计算机, 2021, 2(6): 5-10.

[14] Lin T Y, Maire M, Belongie S, et al. Microsoft COCO: Common objects in context［C］// European Conf. on Computer Vision, 2014.

[15] Andrew A M. Multiple View Geometry in Computer Vision［M］. Cambridge: Kybernetes, 2001.

[16] Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation［C］//Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition, 2015: 3431-3440.

[17] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition［J］. Computer Science, 2014: arXiv: 1409.1556.

[18] Zhang J, Wang C, Liu S, et al. Content-aware unsupervised deep homography estimation［J］. Springer, 2020: arXiv: 1909.05983.

[19] Jaderberg M, Simonyan K, Zisserman A. Spatial transformer networks［J］. Advances in Neural Information Processing Systems, 2015, 28: 2017-2025.

[20] Paszke A, Gross S, Chintala S, et al. Automatic differentiation in PyTorch［C］// 31st Conf. on Neural Information Proc. Systems, 2017.

吴柔莞, 徐智勇, 张建林. 基于无监督级联的亚像素单应矩阵估计[J]. 半导体光电, 2022, 43(1): 158. WU Rouwan, XU Zhiyong, ZHANG Jianlin. Sub-pixel Homography Matrix Estimation Based on Unsupervised Cascade[J]. Semiconductor Optoelectronics, 2022, 43(1): 158.

基于无监督级联的亚像素单应矩阵估计

关于本站 Cookie 的使用提示

全站搜索

基于无监督级联的亚像素单应矩阵估计

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索