Abstract
The paper addresses a new way to reconstruct the 3D shape from a single defocus image using convolutional neutral network (CNN) trained on fractals images. Our method employs point spread function (PSF) with two different degrees of defocus (DoD) to blur a single defocus image and pass them parallelly into CNN to predict the normalized degree of defocus map. Later matting Laplacian interpolation was used to refine the original depth map. Experiments with our own defocus dataset, illustrated that the proposed method achieves accurate shape map with less computing resources. Further network shows great adaptation to real scene defocus images.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Hadsell, R., Sermanet, P., Ben, J., Erkan, A., Lecun, Y.: Learning long-range vision for autonomous off-road driving. J. Field Robot. 26(2), 120–144 (2009)
Lagüela, S., Dorado, I., Gesto, M., Arias, P., Gonzálezaguilera, D., Lorenzo, H.: Behavior analysis of novel wearable indoor mapping system based on 3d-slam. Sensors 18(3), 766 (2018)
Chen, X., Kundu, K., Zhu, Y., Ma, H., Fidler, S., Urtasun, R.: 3d object proposals using stereo imagery for accurate object class detection. IEEE Trans. Pattern Anal. Mach. Intell. PP(99), 1 (2016)
Scharstein, D., Szeliski, R.: A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. Int. J. Comput. Vision 47(1–3), 7–42 (2002)
Oppenheim, A.V., Schafer, R.W.: Digital Signal Processing. Prentice Hall, Englewood Cliffs (1975). ISBN 0-13-214635-5
Ghita, O., Whelan, P.F.: A video-rate range sensor based on depth from defocus. Opt. Laser Technol. 33(3), 167–176 (2001)
Claxton, C.D., Staunton, R.C.: Measurement of the point-spread function of a noisy imaging system. J. Opt. Soc. Am. A: Opt. Image Sci. Vis. 25(1), 159–170 (2008)
Tu, X., Subbarao, M., & Kang, Y.: A new approach to 3D shape recovery of local planar surface patches from shift-variant blurred images. In: International Conference on Pattern Recognition. IEEE (2009)
Rajagopalan, A., Chaudhuri, S.: An MRF model-based approach to simultaneous recovery of depth and restoration from defocused images. IEEE Trans. Pattern Anal. Mach. Intell. 21(7), 577–589 (1999)
Pentland, A.: A new sense for depth of field. IEEE Trans. Pattern Anal. Mach. Intell. 9(4), 523–531 (1987)
Subbarao, M., Surya, G.: Depth from defocus: a spatial domain approach. Int. J. Comput. Vision 13(3), 271–294 (1994)
Watanabe, M., Nayar, S.K.: Telecentric optics for focus analysis. IEEE Trans. Pattern Anal. Mach. Intell. 19(12), 1360–1365 (1997)
Chaudhuri, S., Rajagopalan, A.N., Grafik, D.: Depth From Defocus: A Real Aperture Imaging Approach. Springer, New York (1999)
Hu, H., Haan, G.D.: Low cost robust blur estimator. In: 2006 IEEE International Conference on Image Processing. IEEE (2006)
Zhuo, S., Sim, T.: Defocus map estimation from a single image. Pattern Recogn. 44(9), 1852–1858 (2011)
Levin, A., Lischinski, D., Weiss, Y.: A closed-form solution to natural image matting. IEEE Trans. Pattern Anal. Mach. Intell. 30(2), 228–242 (2007)
Levin, A., Hasinoff, S.W., Green, P., Durand, F., Freeman, W.T.: 4D frequency analysis of computational cameras for depth of field extension. ACM Transactions on Graphics (TOG). ACM (2009)
Liu, F., Shen, C., Lin, G.: Deep convolutional neural fields for depth estimation from a single image. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5162–5170 (2015)
Eigen, D., Puhrsch, C., Fergus, R.: Depth map prediction from a single image using a multi-scale deep network. In: Advances in Neural Information Processing Systems, pp. 2366–2374 (2014)
Anwar, S., Hayder, Z., Porikli, F.: Depth estimation and blur removal from a single out-of-focus image. In: BMVC (2017)
Joseph, K., Raj, A.N.J.: Reconstruction of 3-dimensional scenes using depth from defocus and artificial neural networks trained on fractals. In: International Conference on Soft Computing and Pattern Recognition, pp. 121–129. Springer, Cham, December 2016
Hecht, E.: Optics, 4th edn. Addison Wesley, New York (2001)
Claxton, C.D., Staunton, R.C.: Measurement of the point-spread function of a noisy imaging system. JOSA A 25(1), 159–170 (2008)
Do, M.N., Vetterli, M.: Wavelet-based texture retrieval using generalized Gaussian density and Kullback-Leibler distance. IEEE Trans. Image Process. 11(LCAV-ARTICLE-2002-001), 146–158 (2002)
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)
Mandelbrot, B.B.: The Fractal Geometry of Nature, vol. 1. WH freeman, New York (1982)
Canny, J.: A computational approach to edge detection. IEEE Trans. Pattern Anal. Mach. Intell. 8(6), 679–698 (1986)
Chen, J., Paris, S., Durand, F.: Real-time edge-aware image processing with the bilateral grid. In: ACM Transactions on Graphics (TOG), vol. 26, No. 3, p. 103. ACM, August 2007
Raj, A.N.J., Staunton, R.C.: Rational filter design for depth from defocus. Pattern Recogn. 45(1), 198–207 (2012)
Raj, A.N.J.: Accurate depth from defocus estimation with video-rate implementation (Doctoral dissertation, University of Warwick) (2009)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Chen, R., Joseph Raj, A.N., Ma, X., Zhuang, Z. (2020). Shape Reconstruction from a Monocular Defocus Image Using CNN. In: Bi, Y., Bhatia, R., Kapoor, S. (eds) Intelligent Systems and Applications. IntelliSys 2019. Advances in Intelligent Systems and Computing, vol 1037. Springer, Cham. https://doi.org/10.1007/978-3-030-29516-5_62
Download citation
DOI: https://doi.org/10.1007/978-3-030-29516-5_62
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-29515-8
Online ISBN: 978-3-030-29516-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)