Abstract
The weighted least square (WLS) filter is a popular edge-preserving image smoother that is particularly useful for detail enhancing and HDR tone mapping. However, it suffers from limited edge-preserving capability and high computational cost. Existing deep learning-based filters under the WLS framework are mostly based on supervised learning. They improve the efficiency but not the quality. In this paper, we propose a novel edge-preserving filter under the weighted least square framework based on deep unsupervised learning. According to the spatial-varying smoothing property of the edge-preserving filter, we propose a lightweight fully convolution neural network based on dilated convolutions with varying expanding factors. The proposed filter fully makes use of the 2D neighborhood information, thus it is able to suppress various artifacts. Thanks to the highly optimized framework for deep learning, the proposed filter is highly efficient, enabling the processing of 720P images at interactive rates (\(\approx \)12 fps) on a modern desktop. Experimental results indicate that the proposed filter achieves the state-of-the-art smoothing quality. Therefore, our filter benefits a variety of tasks in the field of image processing and computer graphics.
Similar content being viewed by others
Data Availability
Our source code, trained model and data are available at https://github.com/dtz-dd/deepwls.
References
Zhuang P, Ding X (2020) Underwater image enhancement using an edge-preserving filtering retinex algorithm. Multim Tools Appl 79(25–26):17257–17277. https://doi.org/10.1007/s11042-019-08404-4
Liu K, Li X (2021) De-hazing and enhancement method for underwater and low-light images. Multim Tools Appl 80(13):19421–19439. https://doi.org/10.1007/s11042-021-10740-3
Petschnigg G, Szeliski R, Agrawala M, Cohen MF, Hoppe H, Toyama K (2004) Digital photography with flash and no-flash image pairs. ACM Trans Graph 23(3):664–672. https://doi.org/10.1145/1015706.1015777
Yu S, Li X, Ma M, Zhang X, Chen S (2021) Multi-focus image fusion based on L1 image transform. Multim Tools Appl 80(4):5673–5700. https://doi.org/10.1007/s11042-020-09877-4
Wang X, Jin J, Yang B (2017) Diffusion map based interactive image segmentation. Multim Tools Appl 76(16):17497–17509. https://doi.org/10.1007/s11042-016-4106-7
Kala R, Deepa P (2020) Adaptive fuzzy hexagonal bilateral filter for brain MRI denoising. Multim Tools Appl 79(21–22):15513–15530. https://doi.org/10.1007/s11042-019-7459-x
Zhu S, Gao R, Li Z (2017) Stereo matching algorithm with guided filter and modified dynamic programming. Multim Tools Appl 76(1):199–216. https://doi.org/10.1007/s11042-015-3023-5
Ding X, Chen L, Zheng X, Huang Y, Zeng D (2016) Single image rain and snow removal via guided L0 smoothing filter. Multim Tools Appl 75(5):2697–2712. https://doi.org/10.1007/s11042-015-2657-7
Mondal K, Rabidas R, Dasgupta R (2022) Single image haze removal using contrast limited adaptive histogram equalization based multiscale fusion technique. Multim Tools Appl. https://doi.org/10.1007/s11042-021-11890-0
Tomasi C, Manduchi R (1998) Bilateral filtering for gray and color images. In: International conference on computer vision, pp 839–846 . https://doi.org/10.1109/ICCV.1998.710815
He K, Sun J, Tang X (2013) Guided image filtering. IEEE Trans Pattern Anal Mach Intell 35(6):1397–1409. https://doi.org/10.1109/TPAMI.2012.213
Qiang Z, He L, Chen Y, Chen X, Xu D (2019) Adaptive fast local laplacian filters and its edge-aware application. Multim Tools Appl 78(1):619–639. https://doi.org/10.1007/s11042-017-5347-9
Liu Q, Xiong B, Yang D, Zhang M (2016) A generalized relative total variation method for image smoothing. Multim Tools Appl 75(13):7909–7930. https://doi.org/10.1007/s11042-015-2709-z
Farbman Z, Fattal R, Lischinski D, Szeliski R (2008) Edge-preserving decompositions for multi-scale tone and detail manipulation. ACM Trans Graph 27(3):67–16710. https://doi.org/10.1145/1360612.1360666
Liu W, Zhang P, Huang X, Yang J, Shen C, Reid I (2020) Real-time image smoothing via iterative least squares. ACM Trans Graph 39(3):28–12824. https://doi.org/10.1145/3388887
Yang Y, Zheng H, Zeng L, Shen X, Zhan Y (2022) L1-regularized reconstruction model for edge-preserving filtering. IEEE Trans Multim, 1–1. https://doi.org/10.1109/TMM.2022.3171686
Yang Y, Hui H, Zeng L, Zhao Y, Zhan Y, Yan T (2022) Edge-preserving image filtering based on soft clustering. IEEE Trans Circuits Syst Video Technol 32(7):4150–4162. https://doi.org/10.1109/TCSVT.2021.3124291
Xu L, Lu C, Xu Y, Jia J (2011) Image smoothing via \(L_{0}\) gradient minimization. ACM Trans Graph 30(6):174–117411. https://doi.org/10.1145/2070781.2024208
An X, Pellacini F (2008) Appprop: all-pairs appearance-space edit propagation. ACM Trans Graph 27(3):40. https://doi.org/10.1145/1360612.1360639
Bhat P, Zitnick CL, Cohen MF, Curless B (2010) Gradientshop: A gradient-domain optimization framework for image and video filtering. ACM Trans Graph 29(2):10–11014. https://doi.org/10.1145/1731047.1731048
Lischinski D, Farbman Z, Uyttendaele M, Szeliski R (2006) Interactive local adjustment of tonal values. ACM Trans Graph 25(3):646–653. https://doi.org/10.1145/1141911.1141936
Shao P, Ding S, Ma L, Wu Y, Wu Y (2015) Edge-preserving image decomposition via joint weighted least squares. Comput Vis Media 1(1):37–47. https://doi.org/10.1007/s41095-015-0006-4
Min D, Choi S, Lu J, Ham B, Sohn K, Do MN (2014) Fast global image smoothing based on weighted least squares. IEEE Trans Image Process 23(12):5638–5653. https://doi.org/10.1109/TIP.2014.2366600
Liu W, Chen X, Shen C, Liu Z, Yang J (2017) Semi-global weighted least squares in image filtering. In: International conference on computer vision, pp 5862–5870. https://doi.org/10.1109/ICCV.2017.624
Chen Q, Xu J, Koltun V (2017) Fast image processing with fully-convolutional networks. In: International conference on computer vision, pp 2516–2525. https://doi.org/10.1109/ICCV.2017.273
Wu H, Zheng S, Zhang J, Huang K (2018) Fast end-to-end trainable guided filter. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1838–1847. https://doi.org/10.1109/CVPR.2018.00197
Xu L, Ren J, Yan Q, Liao R, Jia J (2015) Deep edge-aware filters. Proceedings of the 32nd international conference on machine learning. In: Proceedings of machine learning research, 37:1669–1678. https://proceedings.mlr.press/v37/xub15.html
Yin H, Gong Y, Qiu G (2020) Fast and efficient implementation of image filtering using a side window convolutional neural network. Signal Process 176:107717. https://doi.org/10.1016/j.sigpro.2020.107717
Koutis I, Miller GL, Tolliver D (2011) Combinatorial preconditioners and multilevel solvers for problems in computer vision and image processing. Comput Vis Image Underst 115(12):1638–1646. https://doi.org/10.1016/j.cviu.2011.05.013
Krishnan D, Fattal R, Szeliski R (2013) Efficient preconditioning of laplacian matrices for computer graphics. ACM Trans Graph 32(4):142–114215. https://doi.org/10.1145/2461912.2461992
Afonso MV, Bioucas-Dias JM, Figueiredo MAT (2010) Fast image recovery using variable splitting and constrained optimization. IEEE Trans Image Process 19(9):2345–2356. https://doi.org/10.1109/TIP.2010.2047910
Perona P, Malik J (1990) Scale-space and edge detection using anisotropic diffusion. IEEE Trans Pattern Anal Mach Intell 12(7):629–639. https://doi.org/10.1109/34.56205
Barron JT, Poole B (2016) The fast bilateral solver. In: Leibe B, Matas J, Sebe N, Welling M (Eds) Computer vision - ECCV 2016. ECCV 2016. Lecture notes in computer science, vol 9907. Springer, Cham. https://doi.org/10.1007/978-3-319-46487-9_38
Liu W, Zhang P, Chen X, Shen C, Huang X, Yang J (2020) Embedding bilateral filter in least squares for efficient edge-preserving image smoothing. IEEE Trans Circuits Syst Video Technol 30(1):23–35. https://doi.org/10.1109/TCSVT.2018.2890202
Yu F, Koltun V (2016) Multi-scale context aggregation by dilated convolutions. In: Proceedings of the international conference on learning representations
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778. https://doi.org/10.1109/CVPR.2016.90
Kingma DP, Ba J (2015) Adam: A method for stochastic optimization. In: Proceedings of the international conference on learning representations
Everingham M, Gool LV, Williams CKI, Winn JM, Zisserman A (2010) The pascal visual object classes (VOC) challenge. Int J Comput Vis 88(2):303–338. https://doi.org/10.1007/s11263-009-0275-4
Sun Z, Han B, Li J, Zhang J, Gao X (2020) Weighted guided image filtering with steering kernel. IEEE Trans Image Process 29:500–508. https://doi.org/10.1109/TIP.2019.2928631
Xu J, Liu Z, Hou Y, Zhen X, Shao L, Cheng M (2021) Pixel-level non-local image smoothing with objective evaluation. IEEE Trans Multim 23:4065–4078. https://doi.org/10.1109/TMM.2020.3037535
Fan Q, Yang J, Wipf DP, Chen B, Tong X (2018) Image smoothing via unsupervised learning. ACM Trans Graph 37(6):259–125914. https://doi.org/10.1145/3272127.3275081
Zhu F, Liang Z, Jia X, Zhang L, Yu Y (2019) A benchmark for edge-preserving image smoothing. IEEE Trans Image Process 28(7):3556–3570. https://doi.org/10.1109/TIP.2019.2908778
Feng Y, Deng S, Yan X, Yang X, Wei M, Liu L (2021) Easy2hard: Learning to solve the intractables from a synthetic dataset for structure-preserving image smoothing. IEEE Trans Neural Networks Learn Syst, 1–1. https://doi.org/10.1109/TNNLS.2021.3084473
Arbelaez P, Maire M, Fowlkes CC, Malik J (2011) Contour detection and hierarchical image segmentation. IEEE Trans Pattern Anal Mach Intell 33(5):898–916. https://doi.org/10.1109/TPAMI.2010.161
Rother C, Kolmogorov V, Blake A (2004) Grabcut: interactive foreground extraction using iterated graph cuts. ACM Trans Graph 23(3):309–314. https://doi.org/10.1145/1015706.1015720
Ferrari V, Fevrier L, Jurie F, Schmid C (2008) Groups of adjacent contour segments for object detection. IEEE Trans Pattern Anal Mach Intell 30(1):36–51. https://doi.org/10.1109/TPAMI.2007.1144
Girshick RB, Donahue J, Darrell T, Malik J (2014) Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 580–587. https://doi.org/10.1109/CVPR.2014.81
Canny JF (1986) A computational approach to edge detection. IEEE Trans Pattern Anal Mach Intell 8(6):679–698. https://doi.org/10.1109/TPAMI.1986.4767851
Xu L, Yan Q, Xia Y, Jia J (2012) Structure extraction from texture via relative total variation. ACM Trans Graph 31(6):139–113910. https://doi.org/10.1145/2366145.2366158
Karacan L, Erdem E, Erdem A (2013) Structure-preserving image smoothing via region covariances. ACM Trans Graph 32(6):176–117611. https://doi.org/10.1145/2508363.2508403
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Yang, Y., Wu, D., Zeng, L. et al. Weighted least square filter via deep unsupervised learning. Multimed Tools Appl 83, 31361–31377 (2024). https://doi.org/10.1007/s11042-023-16844-2
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-023-16844-2