Abstract
In today’s real-world scenarios’ of computer vision applications, enhancing low-resolution (LR) facial images corrupted with unwanted noise effects is very challenging as the uneven noise distribution severely distorts these images’ local structure. This paper proposes a novel noise-robust face super-resolution (SR) method, namely structural similarity-based Bi-representation SR (SS-BRSR), to tackle this problem. It firstly estimates the true noise level in the corrupted LR face through the novel noise-level estimation algorithm. Afterward, it employs a robust deep-convolutional neural network, namely DnCNN, to separate the pixel-wise noise from the noisy LR face image. This network produces two outputs: (i) a residual image and (ii) a smooth LR face image. We utilize the first output for pixel-wise updating the entire LR training images, making the structural similarity between the test and the training LR images. Further, for SR reconstruction, the SS-BRSR consists of two patch representation components that individually reconstruct the HR faces corresponding to the initial noisy LR and smooth LR face images. Besides, in both the components, the Gradient and Laplacian features-based learning scheme is incorporated to preserve the discriminative facial features in the SR reconstruction. Here, the first component substantially minimizes the reconstruction error due to noise, and the second component compensates for the lost detail in the LR face image. The target HR face image is restored by taking the appropriate proportions of obtained HR face images from each component. The experimental results on different face datasets justify the SS-BRSR method’s superiority over the state-of-the-art face SR methods. For instance, the quantitative performance (in terms of PNSR and SSIM) of the proposed method over the state-of-the-art RLENR and DFDNet methods gained an improvement of [1%, 1.5%, 2.5%, 2.5%] under [10, 15, 20, 30] noise-level densities, and [1%, 1.5%, 2%, 1.5%] under [10, 15, 20, 30] noise-level densities, respectively, for the standard CelebA and FEI datasets.
Similar content being viewed by others
Data Availability
Enquiries about data availability should be directed to the authors.
Notes
Here, the blind level of noise indicates the unknown level of noise.
The complete CelebA dataset is available at: http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html
This dataset is available at: https://github.com/ankushjain01/IIITM-SVI-dataset.git
References
Anwar S, Barnes N (2020) Densely residual laplacian super-resolution. IEEE Trans Pattern Anal Mach Intell
Baker S, Kanade T (2000) Hallucinating faces. In: Proceedings fourth IEEE international conference on automatic face and gesture recognition (Cat. No. PR00580), pp 83–88
Banerjee J, Jawahar CV (2008) Super-resolution of text images using edge-directed tangent field. In: 2008 The eighth IAPR international workshop on document analysis systems, pp 76–83
Chakrabarti A, Rajagopalan AN, Chellappa R (2007) Super-resolution of face images using kernel pca-based prior. IEEE Trans Multimed 9(4):888–892
Chang H, Yeung D-Y, Xiong Y (2004) Super-resolution through neighbor embedding. In: Proceedings of the 2004 IEEE computer society conference on computer vision and pattern recognition, (CVPR), vol 1, pp I275–I282
Dabov K, Foi A, Katkovnik V, Egiazarian K (2007) Image denoising by sparse 3-d transform-domain collaborative filtering. IEEE Trans Image Process 16(8):2080–2095
Dong C, Loy CC, He K, Tang X (2016) Image super-resolution using deep convolutional networks. IEEE Trans Pattern Anal Mach Intell 38 (2):295–307
Dong C, Zhu X, Deng Y, Loy CC, Qiao Y (2015) Boosting optical character recognition: a super-resolution approach, CoRR arXiv:1506.02211
Farrugia RA, Guillemot C (2017) Face hallucination using linear models of coupled sparse support. IEEE Trans Image Process 26(9):4562–4577
Farsiu S, Robinson MD, Elad M, Milanfar P (2004) Fast and robust multiframe super resolution. IEEE Trans Image Process 13(10):1327–1344
Freeman WT, Pasztor EC, Carmichael OT (2000) Learning low-level vision. Int J Comput Vis 40(1):25–47
Gao G, Yang J (2014) A novel sparse representation based framework for face image super-resolution. Neurocomputing 134:92–99
Greenspan H (2009) Super-resolution in medical imaging. Comput J 52(1):43–63
Guo S, Yan Z, Zhang K, Zuo W, Zhang L (2019) Toward convolutional blind denoising of real photographs. In: 2019 IEEE/CVF conference on computer vision and pattern recognition (CVPR), pp 1712–1722
Gupta S, Thakur K, Kumar M (2021) 2d-Human face recognition using sift and surf descriptors of face’s feature regions. Vis Comput 37(3):447–456
Huang H, He R, Sun Z, Tan T (2017) Wavelet-srnet: a wavelet-based cnn for multi-scale face super resolution. In: IEEE international conference on computer vision, pp 1689–1697
Jia Z, Wang H, Xiong Z, Finn A (2011) Fast face hallucination with sparse representation for video surveillance. In: The first asian conference on pattern recognition, pp 179–183
Jiang J, Chen C, Huang K, Cai Z, Hu R (2016) Noise robust position-patch based face super-resolution via tikhonov regularized neighbor representation. Inf Sci 367-368:354–372
Jiang J, Hu R, Wang Z, Han Z (2014) Noise robust face hallucination via locality-constrained representation. IEEE Trans Multimed 16(5):1268–1281
Jiang J, Hu R, Wang Z, Han Z (2014) Face super-resolution via multilayer locality-constrained iterative neighbor embedding and intermediate dictionary learning. IEEE Trans Image Process 23(10):4220–4231
Jiang J, Ma J, Chen C, Jiang X, Wang Z (2017) Noise robust face image super-resolution through smooth sparse representation. IEEE Trans Cybern 47(11):3991–4002
Jiang J, Yu Y, Tang S, Ma J, Aizawa A, Aizawa K (2018) Context-patch face hallucination based on thresholding locality-constrained representation and reproducing learning. IEEE Trans Cybern:1–14
Jiao Q, Zhong J, Liu C, Wu S, Wong H-S (2022) Perturbation-insensitive cross-domain image enhancement for low-quality face verification. Inf Sci 608:1183–1201
Jin D, Bai X (2019) Patch-sparsity-based image inpainting through a facet deduced directional derivative. IEEE Trans Circuits Syst Video Technol 29(5):1310–1324
Jung C, Jiao L, Liu B, Gong M (2011) Position-patch based face hallucination using convex optimization. IEEE Signal Process 18(6):367–370
Kim J, Lee JK, Lee KM (2016) Accurate image super-resolution using very deep convolutional networks. In: 2016 IEEE conference on computer vision and pattern recognition (CVPR), pp 1646–1654
Kouame D, Ploquin M (2009) Super-resolution in medical imaging : an illustrative approach through ultrasound. In: 2009 IEEE international symposium on biomedical imaging: from Nano to Macro, pp 249–252
Kumar M, Jindal MK, Kumar M (2022) Design of innovative captcha for hindi language. Neural Comput Appl 34(6):4957–4992
Kumar A, Kaur A, Kumar M (2019) Face detection techniques: a review. Artif Intell Rev
Kumar A, Kumar M, Kaur A (2021) Face detection in still images under occlusion and non-uniform illumination. Multimed Tools Appl 80(10):14565–14590
Kumar K, Shrimankar DD (2018) F-des: fast and deep event summarization. IEEE Trans Multimed 20(2):323–334
Li X, Chen C, Zhou S, Lin X, Zuo W, Zhang L (2020) Blind face restoration via deep multi-scale component dictionaries. In: European conference on computer vision, Springer, pp 399–415
Lin FC, Fookes CB, Chandran V, Sridharan S (2005) Investigation into optical flow super-resolution for surveillance applications. In: APRS workshop on digital image computing: pattern recognition and imaging for medical applications, pp 73–78
Liu L, Chen CLP, Li S, Tang YY, Chen L (2018) Robust face hallucination via locality-constrained bi-layer representation. IEEE Trans Cybern 48 (4):1189–1201
Liu L, Liu H, Li S, Chen CP (2020) Face hallucination via multiple feature learning with hierarchical structure. Inf Sci 512:416–430
Liu L, Liu H, Li S, Chen CLP (2020) Face hallucination via multiple feature learning with hierarchical structure. Inf Sci 512:416–430
Liu Z, Luo P, Wang X, Tang X (2015) Deep learning face attributes in the wild. In: Proceedings of international conference on computer vision (ICCV)
Liu X, Tanaka M, Okutomi M (2013) Single-image noise level estimation for blind denoising. IEEE Trans Image Process 22(12):5226–5237
Liu L, Tang X, Chen CP, Cai L, Lan R (2022) Superpixel-guided locality quaternion representation for color face hallucination. Inf Sci 609:565–577
Ma X, Zhang J, Qi C (2010) Hallucinating face by position-patch. Pattern Recognit 43(6):2224–2236
Makantasis K, Doulamis AD, Doulamis ND, Nikitakis A (2018) Tensor-based classification models for hyperspectral data analysis. IEEE Trans Geosci Remote Sens 56(12):6884–6898
Nagar S, Jain A, Singh PK, Kumar A (2020) Pixel-wise dictionary learning based locality-constrained representation for noise robust face hallucination. Digital Signal Process 99:102667
Nagar S, Jain A, Singh PK, Kumar A (2021) Mixed-noise robust face super-resolution through residual-learning based error suppressed nearest neighbor representation. Inf Sci 546:121–145
Nagar S, Jain A, Singh PK, Kumar A (2022) Adaptive optimal multi-features learning based representation for face hallucination. Expert Syst Appl 190:116141
Negi A, Kumar K (2021) Face mask detection in real-time video stream using deep learning. Computat Intell Healthcare Inf:255–268
Negi A, Kumar K, Chaudhari NS, Singh N, Chauhan P (2021) Predictive analytics for recognizing human activities using residual network and fine-tuning. In: International conference on big data analytics, Springer, pp 296–310
Negi A, Kumar K, Chauhan P, Rajput R (2021) Deep neural architecture for face mask detection on simulated masked face dataset against covid-19 pandemic. In: 2021 International conference on computing, communication, and intelligent systems (ICCCIS), pp 595–600
Park J, Lee S (2008) An example-based face hallucination method for single-frame, low-resolution facial images. IEEE Trans Image Process 17(10):1806–1816
Pei X, Guan Y, Cai P, Dong T (2018) Face hallucination via gradient constrained sparse representation. IEEE Access 6:4577–4586
Roweis ST, Saul LK (2000) Nonlinear dimensionality reduction by locally linear embedding. Science 290(5500):2323–2326
Rowley HA, Baluja S, Kanade T (1998) Neural network-based face detection. IEEE Trans Pattern Anal Mach Intell 20(1):23–38
Sharma S, Kumar K (2021) Asl-3dcnn: american sign language recognition technique using 3-d convolutional neural networks. Multimed Tools Appl 80(17):26319–26331
Sharma S, Kumar K, Singh N (2017) D-fes: deep facial expression recognition system. In: 2017 Conference on information and communication technology (CICT), pp 1–6
Shi J, Liu X, Qi C (2014) Global consistency, local sparsity and pixel correlation: a unified framework for face hallucination. Pattern Recognit 47(11):3520–3534
Shi J, Qi C (2015) From local geometry to global structure: Learning latent subspace for low-resolution face image recognition. IEEE Signal Process Lett 22(5):554–558
Singh S, Ahuja U, Kumar M, Kumar K, Sachdeva M (2021) Face mask detection using yolov3 and faster r-cnn models: covid-19 environment. Multimed Tools Appl 80(13):19753–19768
Singh A, Porikli F, Ahuja N (2014) Super-resolving noisy images. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2846–2853
Thomaz CE, Giraldi GA (2010) A new ranking method for principal components analysis and its application to face image analysis. Image Vis Comput 28 (6):902–913
Wang Z, Bovik AC, Sheikh HR, Simoncelli EP (2004) Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process 13(4):600–612
Wang N, Gao X, Sun L, Li J (2017) Bayesian face sketch synthesis. IEEE Trans Image Process 26(3):1264–1274
Wang N, Tao D, Gao X, Li X, Li J (2014) A comprehensive survey to face hallucination. Int J Comput Vis 106(1):9–30
Wang J, Yang J, Yu K, Lv F, Huang T, Gong Y (2010) Locality-constrained linear coding for image classification. In: 2010 IEEE computer society conference on computer vision and pattern recognition, pp 3360–3367
Wei S, Zhou X, Wu W, Pu Q, Wang Q, Yang X (2018) Medical image super-resolution by using multi-dictionary and random forest. Sustainable Cities Society 37:358–370
Xu S, Zeng X, Jiang Y, Tang Y (2017) A multiple image-based noise level estimation algorithm. IEEE Signal Process Lett 24(11):1701–1705
Yang S, Liu J, Fang Y, Guo Z (2018) Joint-feature guided depth map super-resolution with face priors. IEEE Trans Cybern 48(1):399–411
Yang J, Wright J, Huang TS, Ma Y (2010) Image super-resolution via sparse representation. IEEE Trans Image Process 19(11):2861–2873
Yu X, Shiri F, Ghanem B, Porikli F (2020) Can we see more? joint frontalization and hallucination of unaligned tiny faces. IEEE Trans Pattern Anal Mach Intell 42(9):2148–2164
Yu K, Zhang T, Gong Y (2009) Nonlinear learning using local coordinate coding. In: Proceedings of the 22nd international conference on neural information processing systems, NIPS’09, Curran associates inc., pp 2223–2231
Zhang Y, Li K, Li K, Wang L, Zhong B, Fu Y (2018) Image super-resolution using very deep residual channel attention networks. In: Proceedings of the European conference on computer vision (ECCV), pp 286–301
Zhang Y, Tsang IW, Li J, Liu P, Lu X, Yu X (2021) Face hallucination with finishing touches. IEEE Trans Image Process 30:1728–1743
Zhang J, Yu J, Tao D (2018) Local deep-feature alignment for unsupervised dimension reduction. IEEE Trans Image Process 27(5):2420–2432
Zhang K, Zuo W, Chen Y, Meng D, Zhang L (2017) Beyond a gaussian denoiser: residual learning of deep cnn for image denoising. IEEE Trans Image Process 26(7):3142–3155
Zou WWW, Yuen PC (2012) Very low resolution face recognition problem. IEEE Trans Image Process 21(1):327–340
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of Interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix A: Algorithm for LR face representation through multi-feature learning
Appendix B: Optimization for noisy LR face representation through the residual updated LR training face images samples and multi-feature learning (Objection function (8))
The objective function given in (8) can be written as the following matrix arrangement.
Further, for sake of simplicity, we assume (p,q)th position patch as ith patch; hence, (B.1) is reformulated as follows.
Here, \( y_{LR_{Res}}^{k,i} \) is a matrix having column-wise K1-NN training patches of ith LR test patch, \( \mathcal {F}_{j}(y_{LR_{Res}}^{k,i}) \) denotes a matrix having column-wise features of respective K1-NN training patches corresponding to feature index \( j \in \{1,2,\dots ,F\} \), and \( \mathcal {D}_{1} \) is the K1 × K1 diagonal matrix defined as follows.
By introducing the following auxiliary variables,
Equation (B.2) can be transformed into the following.
Equation (B.5) is an example of the regularized-least-square problem and its solution can be analytically derived as follows.
Here, 1 is a N × 1 column vector of ones, Zi denotes the covariance matrix for \( Q^{i}_{LR} \) and it is expressed as follows.
The final optimal weight is obtained by re-scaling it to satisfy the constraint \({\sum }_{k=1}^{K_{1}}w_{k}=1\).
Appendix C: Visual and quantitative results on CelebA dataset
Appendix D: Visual results on real-world ABV-IIITM and CMU-MIT surveillance datasets
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
Nagar, S., Jain, A., Singh, P.K. et al. Structural similarity-based Bi-representation through true noise level for noise-robust face super-resolution. Multimed Tools Appl 82, 26255–26288 (2023). https://doi.org/10.1007/s11042-022-14325-6
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-022-14325-6