Learning Enriched Features for Real Image Restoration and Enhancement

Cited by: 1|Bibtex|Views14|Links
Keywords:
image denoisingimage super resolutionsingle image super resolutionimage restoration taskrecursive residual groupsMore(13+)
Weibo:
We propose a novel architecture whose main branch is dedicated to full-resolution processing and the complementary set of parallel branches provides better contextualized features

Abstract:

With the goal of recovering high-quality image content from its degraded version, image restoration enjoys numerous applications, such as in surveillance, computational photography, medical imaging, and remote sensing. Recently, convolutional neural networks (CNNs) have achieved dramatic improvements over conventional approaches for ima...More
0
Introduction
  • Image content is exponentially growing due to the ubiquitous presence of cameras on various devices.
  • Degradations of different severities are often introduced.
  • It is either because of the physical limitations of cameras, or due to inappropriate lighting conditions.
  • Smart phone cameras come with narrow aperture, and have small sensors with limited dynamic range.
  • They frequently generate noisy and low-contrast images.
  • It is an ill-posed inverse problem, due to the existence of many possible solutions
Highlights
  • Image content is exponentially growing due to the ubiquitous presence of cameras on various devices
  • We propose a new approach capable of performing image denoising, super-resolution and image enhancement
  • We provide details of the multi-scale residual block, which is the fundamental building block of our method, containing several key elements: (a) parallel multi-resolution convolution streams for extracting semantically-richer and spatially-precise feature representations, (b) information exchange across multi-resolution streams, (c) attention-based aggregation of features arriving from multiple streams, (d) dual-attention units to capture contextual information in both spatial and channel dimensions, and (e) residual resizing modules to perform downsampling and upsampling operations
  • We propose a novel architecture whose main branch is dedicated to full-resolution processing and the complementary set of parallel branches provides better contextualized features
  • We propose novel mechanisms to learn relationships between features within each branch as well as across multiscale branches
  • Our feature fusion strategy ensures that the receptive field can be dynamically adapted without sacrificing the original feature details
Methods
  • The authors first present an overview of the proposed MIRNet for image restoration and enhancement, illustrated in Fig. 1.
  • BM3D FoE WNNM KSVD MCWNNM FFDNet+ TWSC CBDNet RIDNet VDN MIRNet [126] [19] [11] [20] [84] [37] [2] [111] [119] [110] [38] 26.90 dB Noisy.
  • The authors analyze the feature aggregation strategy in Table 8
  • It shows that the proposed SKFF generates favorable results compared to summation and concatenation.
  • In Table 9 the authors study how the number of convolutional streams and columns (DAU blocks) of MRB affect the image restoration quality.
  • Additional ablation studies and qualitative results are provided in the supplementary material
Results
  • It is worth noting that CBDNet [38] and RIDNet [5] use additional training data, yet the method provides significantly better results.
Conclusion
  • Concluding Remarks

    Conventional image restoration and enhancement pipelines either stick to the full resolution features along the network hierarchy or use an encoder-decoder architecture.
  • The first approach helps retain precise spatial details, while the latter one provides better contextualized representations.
  • These methods can satisfy only one of the above two requirements, real-world image restoration tasks demand a combination of both conditioned on the given input sample.
  • The authors propose a novel architecture whose main branch is dedicated to full-resolution processing and the complementary set of parallel branches provides better contextualized features.
  • Consistent achievement of state-of-the-art results on five datasets for three image restoration and enhancement tasks corroborates the effectiveness of the approach
Summary
  • Introduction:

    Image content is exponentially growing due to the ubiquitous presence of cameras on various devices.
  • Degradations of different severities are often introduced.
  • It is either because of the physical limitations of cameras, or due to inappropriate lighting conditions.
  • Smart phone cameras come with narrow aperture, and have small sensors with limited dynamic range.
  • They frequently generate noisy and low-contrast images.
  • It is an ill-posed inverse problem, due to the existence of many possible solutions
  • Methods:

    The authors first present an overview of the proposed MIRNet for image restoration and enhancement, illustrated in Fig. 1.
  • BM3D FoE WNNM KSVD MCWNNM FFDNet+ TWSC CBDNet RIDNet VDN MIRNet [126] [19] [11] [20] [84] [37] [2] [111] [119] [110] [38] 26.90 dB Noisy.
  • The authors analyze the feature aggregation strategy in Table 8
  • It shows that the proposed SKFF generates favorable results compared to summation and concatenation.
  • In Table 9 the authors study how the number of convolutional streams and columns (DAU blocks) of MRB affect the image restoration quality.
  • Additional ablation studies and qualitative results are provided in the supplementary material
  • Results:

    It is worth noting that CBDNet [38] and RIDNet [5] use additional training data, yet the method provides significantly better results.
  • Conclusion:

    Concluding Remarks

    Conventional image restoration and enhancement pipelines either stick to the full resolution features along the network hierarchy or use an encoder-decoder architecture.
  • The first approach helps retain precise spatial details, while the latter one provides better contextualized representations.
  • These methods can satisfy only one of the above two requirements, real-world image restoration tasks demand a combination of both conditioned on the given input sample.
  • The authors propose a novel architecture whose main branch is dedicated to full-resolution processing and the complementary set of parallel branches provides better contextualized features.
  • Consistent achievement of state-of-the-art results on five datasets for three image restoration and enhancement tasks corroborates the effectiveness of the approach
Tables
  • Table1: Denoising comparisons on the SIDD dataset [<a class="ref-link" id="c1" href="#r1">1</a>]
  • Table2: Denoising comparisons on the DND dataset [<a class="ref-link" id="c78" href="#r78">78</a>]
  • Table3: Super-resolution evaluation on the RealSR [<a class="ref-link" id="c13" href="#r13">13</a>] dataset. Compared to the state-of-the-art, our method consistently yields significantly better image quality scores for all three scaling factors
  • Table4: Cross-camera generalization test for super-resolution. Networks trained for one camera are tested on the other camera. Our MIRNet shows good generalization for all possible cases
  • Table5: Low-light image enhancement evaluation on the LoL dataset [<a class="ref-link" id="c106" href="#r106">106</a>]. The proposed method significantly advances the state-of-the-art
  • Table6: Image enhancement comparisons on the MIT-Adobe FiveK dataset [<a class="ref-link" id="c12" href="#r12">12</a>]
  • Table7: Impact of individual components of MRB
  • Table8: Feature aggregation. Our SKFF uses ∼ 6× fewer parameters than concat, but generates better results
  • Table9: Ablation study on different layouts of MRB. Rows denote number of parallel resolution streams, and Cols represent the number of columns containing DAUs
Download tables as Excel
Related work
  • With the rapidly growing image media content, there is a pressing need to develop effective image restoration and enhancement algorithms. In this paper, we propose a new approach capable of performing image denoising, super-resolution and image enhancement. Different from existing works for these problems, our approach processes features at the original resolution in order to preserve spatial details, while effectively fuses contextual information from multiple parallel branches. Next, we briefly describe the representative methods for each of the studied problems.

    Image denoising. Classic denoising methods are mainly based on modifying transform coefficients [114,29,89] or averaging neighborhood pixels [90,97,77,85]. Although the classical methods perform well, the self-similarity [30] based algorithms, e.g., NLM [10] and BM3D [20], demonstrate promising denoising performance. Numerous patch-based algorithms that exploit redundancy (selfsimilarity) in images are later developed [27,37,69,42]. Recently, deep learningbased approaches [11,5,9,34,38,79,118,119] make significant advances in image denoising, yielding favorable results than those of the hand-crafted methods.
Reference
  • Abdelhamed, A., Lin, S., Brown, M.S.: A high-quality denoising dataset for smartphone cameras. In: CVPR (2018) 8, 9, 10
    Google ScholarFindings
  • Aharon, M., Elad, M., Bruckstein, A.: K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation. Trans. Sig. Proc. (2006) 9
    Google ScholarLocate open access versionFindings
  • Ahn, N., Kang, B., Sohn, K.A.: Fast, accurate, and lightweight super-resolution with cascading residual network. In: ECCV (2018) 4
    Google ScholarFindings
  • Allebach, J., Wong, P.W.: Edge-directed interpolation. In: ICIP (1996) 3 5. Anwar, S., Barnes, N.: Real image denoising with feature attention. ICCV (2019)
    Google ScholarLocate open access versionFindings
  • arXiv (2019) 3, 10 7.
    Google ScholarFindings
  • Badrinarayanan, V., Kendall, A., Cipolla, R.: SegNet: a deep convolutional encoder-decoder architecture for image segmentation. TPAMI (2017) 5 8.
    Google ScholarLocate open access versionFindings
  • Bertalmıo, M., Caselles, V., Provenzi, E., Rizzi, A.: Perceptual color correction through variational techniques. TIP (2007) 4 9.
    Google ScholarLocate open access versionFindings
  • Brooks, T., Mildenhall, B., Xue, T., Chen, J., Sharlet, D., Barron, J.T.: Unprocessing images for learned raw denoising. In: CVPR (2019) 3 10.
    Google ScholarFindings
  • Buades, A., Coll, B., Morel, J.M.: A non-local algorithm for image denoising. In: CVPR (2005) 3, 9 11.
    Google ScholarFindings
  • Burger, H.C., Schuler, C.J., Harmeling, S.: Image denoising: Can plain neural networks compete with BM3D? In: CVPR (2012) 3, 9 12.
    Google ScholarFindings
  • Bychkovsky, V., Paris, S., Chan, E., Durand, F.: Learning photographic global tonal adjustment with a database of input/output image pairs. In: CVPR (2011) 8, 12, 13 13.
    Google ScholarFindings
  • Cai, J., Zeng, H., Yong, H., Cao, Z., Zhang, L.: Toward real-world single image super-resolution: A new benchmark and a new model. In: ICCV (2019) 8, 10, 11, 12 14.
    Google ScholarFindings
  • Chang, H., Yeung, D.Y., Xiong, Y.: Super-resolution through neighbor embedding. In: CVPR (2004) 3 15.
    Google ScholarFindings
  • Charbonnier, P., Blanc-Feraud, L., Aubert, G., Barlaud, M.: Two deterministic half-quadratic regularization algorithms for computed imaging. In: ICIP (1994) 5 16.
    Google ScholarFindings
  • Chen, C., Chen, Q., Xu, J., Koltun, V.: Learning to see in the dark. In: CVPR (2018) 2 17.
    Google ScholarFindings
  • Chen, L.C., Zhu, Y., Papandreou, G., Schroff, F., Adam, H.: Encoder-decoder with atrous separable convolution for semantic image segmentation. In: ECCV (2018) 4 18.
    Google ScholarFindings
  • Chen, Y.S., Wang, Y.C., Kao, M.H., Chuang, Y.Y.: Deep photo enhancer: Unpaired learning for image enhancement from photographs with gans. In: CVPR (2018) 4, 12, 13 19.
    Google ScholarFindings
  • Chen, Y., Yu, W., Pock, T.: On learning optimized reaction diffusion processes for effective image restoration. In: CVPR (2015) 9 20.
    Google ScholarFindings
  • Dabov, K., Foi, A., Katkovnik, V., Egiazarian, K.: Image denoising by sparse 3-D transform-domain collaborative filtering. TIP (2007) 3, 9 21.
    Google ScholarLocate open access versionFindings
  • Dahl, R., Norouzi, M., Shlens, J.: Pixel recursive super resolution. In: ICCV (2017) 4 22.
    Google ScholarFindings
  • Dai, T., Cai, J., Zhang, Y., Xia, S.T., Zhang, L.: Second-order attention network for single image super-resolution. In: CVPR (2019) 4, 6
    Google ScholarFindings
  • 23. https://noise.visinf.tu-darmstadt.de/benchmark/ (2017), [Online; accessed 29-Feb-2020] 8
    Findings
  • 24. Deng, Y., Loy, C.C., Tang, X.: Aesthetic-driven image enhancement by adversarial learning. In: ACM Multimedia (2018) 4
    Google ScholarLocate open access versionFindings
  • 25. Dong, C., Loy, C.C., He, K., Tang, X.: Learning a deep convolutional network for image super-resolution. In: ECCV (2014) 3
    Google ScholarFindings
  • 26. Dong, C., Loy, C.C., He, K., Tang, X.: Image super-resolution using deep convolutional networks. TPAMI (2015) 2, 3
    Google ScholarLocate open access versionFindings
  • 27. Dong, W., Shi, G., Li, X.: Nonlocal image restoration with bilateral variance estimation: a low-rank approach. TIP (2012) 3
    Google ScholarLocate open access versionFindings
  • 28. Dong, X., Wang, G., Pang, Y., Li, W., Wen, J., Meng, W., Lu, Y.: Fast efficient algorithm for enhancement of low lighting video. In: ICME (2011) 12
    Google ScholarFindings
  • 29. Donoho, D.L.: De-noising by soft-thresholding. Trans. on information theory (1995) 3
    Google ScholarFindings
  • 30. Efros, A.A., Leung, T.K.: Texture synthesis by non-parametric sampling. In: ICCV (1999) 3
    Google ScholarFindings
  • 31. Fourure, D., Emonet, R., Fromont, E., Muselet, D., Tremeau, A., Wolf, C.: Residual conv-deconv grid network for semantic segmentation. In: BMVC (2017) 5
    Google ScholarFindings
  • 32. Freedman, G., Fattal, R.: Image and video upscaling from local self-examples. TOG (2011) 3
    Google ScholarLocate open access versionFindings
  • 33. Fu, X., Zeng, D., Huang, Y., Zhang, X.P., Ding, X.: A weighted variational model for simultaneous reflectance and illumination estimation. In: CVPR (2016) 12, 13
    Google ScholarFindings
  • 34. Gharbi, M., Chaurasia, G., Paris, S., Durand, F.: Deep joint demosaicking and denoising. TOG (2016) 3
    Google ScholarLocate open access versionFindings
  • 35. Gharbi, M., Chen, J., Barron, J.T., Hasinoff, S.W., Durand, F.: Deep bilateral learning for real-time image enhancement. TOG (2017) 12, 13
    Google ScholarLocate open access versionFindings
  • 36. Gu, S., Li, Y., Gool, L.V., Timofte, R.: Self-guided network for fast image denoising. In: ICCV (2019) 2
    Google ScholarFindings
  • 37. Gu, S., Zhang, L., Zuo, W., Feng, X.: Weighted nuclear norm minimization with application to image denoising. In: CVPR (2014) 3, 9
    Google ScholarFindings
  • 38. Guo, S., Yan, Z., Zhang, K., Zuo, W., Zhang, L.: Toward convolutional blind denoising of real photographs. In: CVPR (2019) 3, 9, 10
    Google ScholarFindings
  • 39. Guo, X., Li, Y., Ling, H.: Lime: Low-light image enhancement via illumination map estimation. TIP (2016) 12, 13
    Google ScholarLocate open access versionFindings
  • 40. Han, W., Chang, S., Liu, D., Yu, M., Witbrock, M., Huang, T.S.: Image superresolution via dual-state recurrent networks. In: CVPR (2018) 4
    Google ScholarFindings
  • 41. He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR (2016) 3
    Google ScholarFindings
  • 42. Hedjam, R., Moghaddam, R.F., Cheriet, M.: Markovian clustering for the nonlocal means image denoising. In: ICIP (2009) 3
    Google ScholarFindings
  • 43. Hu, J., Shen, L., Sun, G.: Squeeze-and-excitation networks. In: CVPR (2018) 6 44. Hu, Y., He, H., Xu, C., Wang, B., Lin, S.: Exposure: A white-box photo postprocessing framework. TOG (2018) 8, 12 45.
    Google ScholarLocate open access versionFindings
  • Huang, G., Chen, D., Li, T., Wu, F., van der Maaten, L., Weinberger, K.Q.: Multi-scale dense networks for resource efficient image classification. In: ICLR (2018) 5 46.
    Google ScholarFindings
  • Hubel, D.H., Wiesel, T.N.: Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex. The Journal of physiology (1962) 5 47.
    Google ScholarLocate open access versionFindings
  • Hui, Z., Wang, X., Gao, X.: Fast and accurate single image super-resolution via information distillation network. In: CVPR (2018) 3
    Google ScholarFindings
  • 48. Hung, C.P., Kreiman, G., Poggio, T., DiCarlo, J.J.: Fast readout of object identity from macaque inferior temporal cortex. Science (2005) 5
    Google ScholarLocate open access versionFindings
  • 49. Ignatov, A., Kobyshev, N., Timofte, R., Vanhoey, K., Van Gool, L.: Dslr-quality photos on mobile devices with deep convolutional networks. In: ICCV (2017) 2
    Google ScholarFindings
  • 50. Ignatov, A., Kobyshev, N., Timofte, R., Vanhoey, K., Van Gool, L.: Wespe: weakly supervised photo enhancer for digital cameras. In: CVPRW (2018) 4
    Google ScholarFindings
  • 51. Irani, M., Peleg, S.: Improving resolution by image registration. CVGIP (1991) 3 52. Jobson, D.J., Rahman, Z.u., Woodell, G.A.: A multiscale retinex for bridging the gap between color images and the human observation of scenes. TIP (1997) 4, 12 53.
    Google ScholarLocate open access versionFindings
  • (1981) 3, 10 54.
    Google ScholarFindings
  • Kim, J., Kwon Lee, J., Mu Lee, K.: Accurate image super-resolution using very deep convolutional networks. In: ICCV (2016) 3, 10, 11 55.
    Google ScholarFindings
  • Kim, J., Kwon Lee, J., Mu Lee, K.: Deeply-recursive convolutional network for image super-resolution. In: CVPR (2016) 4 56.
    Google ScholarLocate open access versionFindings
  • Kim, K.I., Kwon, Y.: Single-image super-resolution using sparse regression and natural image prior. TPAMI (2010) 3 57.
    Google ScholarFindings
  • Kupyn, O., Martyniuk, T., Wu, J., Wang, Z.: Deblurgan-v2: Deblurring (ordersof-magnitude) faster and better. In: ICCV (2019) 2 58.
    Google ScholarFindings
  • Lai, W.S., Huang, J.B., Ahuja, N., Yang, M.H.: Deep laplacian pyramid networks for fast and accurate superresolution. In: CVPR (2017) 4 59.
    Google ScholarLocate open access versionFindings
  • Land, E.H.: The retinex theory of color vision. Scientific american (1977) 4 60.
    Google ScholarLocate open access versionFindings
  • Lebrun, M., Colom, M., Morel, J.M.: The noise clinic: a blind image denoising algorithm. IPOL (2015) 9 61.
    Google ScholarFindings
  • Ledig, C., Theis, L., Huszar, F., Caballero, J., Cunningham, A., Acosta, A., Aitken, A., Tejani, A., Totz, J., Wang, Z., et al.: Photo-realistic single image super-resolution using a generative adversarial network. In: CVPR (2017) 4, 10, 11 62.
    Google ScholarFindings
  • Li, J., Fang, F., Mei, K., Zhang, G.: Multi-scale residual network for image superresolution. In: ECCV (2018) 4 63.
    Google ScholarFindings
  • Li, X., Wang, W., Hu, X., Yang, J.: Selective kernel networks. In: CVPR (2019) 5 64.
    Google ScholarLocate open access versionFindings
  • Li, X., Orchard, M.T.: New edge-directed interpolation. TIP (2001) 3 65.
    Google ScholarFindings
  • Lim, B., Son, S., Kim, H., Nah, S., Mu Lee, K.: Enhanced deep residual networks for single image super-resolution. In: CVPRW (2017) 4 66.
    Google ScholarFindings
  • Liu, Y., Wang, R., Shan, S., Chen, X.: Structure inference net: Object detection using scene-level context and instance-level relationships. In: CVPR (2018) 12 67.
    Google ScholarLocate open access versionFindings
  • Lore, K.G., Akintayo, A., Sarkar, S.: LLNet: a deep autoencoder approach to natural low-light image enhancement. Pattern Recognition (2017) 4 68.
    Google ScholarFindings
  • Loshchilov, I., Hutter, F.: Sgdr: Stochastic gradient descent with warm restarts. In: ICLR (2017) 9 69.
    Google ScholarFindings
  • Mairal, J., Bach, F., Ponce, J., Sapiro, G., Zisserman, A.: Non-local sparse models for image restoration. In: ICCV (2009) 3 70.
    Google ScholarFindings
  • Nah, S., Kim, T.H., Lee, K.M.: Deep multi-scale convolutional neural network for dynamic scene deblurring. In: CVPR (2017) 2 71.
    Google ScholarFindings
  • Newell, A., Yang, K., Deng, J.: Stacked hourglass networks for human pose estimation. In: ECCV (2016) 5 72.
    Google ScholarFindings
  • Noh, H., Hong, S., Han, B.: Learning deconvolution network for semantic segmentation. In: ICCV (2015) 5 73.
    Google ScholarLocate open access versionFindings
  • Palma-Amestoy, R., Provenzi, E., Bertalmıo, M., Caselles, V.: A perceptually inspired variational framework for color enhancement. TPAMI (2009) 4
    Google ScholarFindings
  • 74. Park, J., Lee, J.Y., Yoo, D., So Kweon, I.: Distort-and-recover: Color enhancement using deep reinforcement learning. In: CVPR (2018) 8, 12
    Google ScholarFindings
  • 75. Park, S.J., Son, H., Cho, S., Hong, K.S., Lee, S.: SRFEAT: Single image superresolution with feature discrimination. In: ECCV (2018) 4
    Google ScholarFindings
  • 76. Peng, X., Feris, R.S., Wang, X., Metaxas, D.N.: A recurrent encoder-decoder network for sequential face alignment. In: ECCV (2016) 5
    Google ScholarLocate open access versionFindings
  • 77. Perona, P., Malik, J.: Scale-space and edge detection using anisotropic diffusion. TPAMI (1990) 3
    Google ScholarFindings
  • 78. Plotz, T., Roth, S.: Benchmarking denoising algorithms with real photographs. In: CVPR (2017) 8, 9
    Google ScholarLocate open access versionFindings
  • 79. Plotz, T., Roth, S.: Neural nearest neighbors networks. In: NeurIPS (2018) 3 80. Ren, W., Liu, S., Ma, L., Xu, Q., Xu, X., Cao, X., Du, J., Yang, M.H.: Low-light image enhancement via a deep hybrid network. TIP (2019) 4 81.
    Google ScholarFindings
  • Nature neuroscience (1999) 5 82.
    Google ScholarLocate open access versionFindings
  • Rizzi, A., Gatta, C., Marini, D.: From retinex to automatic color equalization: issues in developing a new algorithm for unsupervised color equalization. Journal of Electronic Imaging (2004) 4 83.
    Google ScholarFindings
  • Ronneberger, O., Fischer, P., Brox, T.: U-net: convolutional networks for biomedical image segmentation. In: MICCAI (2015) 2, 5 84.
    Google ScholarLocate open access versionFindings
  • Roth, S., Black, M.J.: Fields of experts. IJCV (2009) 9 85.
    Google ScholarFindings
  • Rudin, L.I., Osher, S., Fatemi, E.: Nonlinear total variation based noise removal algorithms. Physica D: nonlinear phenomena (1992) 3 86.
    Google ScholarFindings
  • Sajjadi, M.S., Scholkopf, B., Hirsch, M.: Enhancenet: Single image superresolution through automated texture synthesis. In: ICCV (2017) 4 87.
    Google ScholarLocate open access versionFindings
  • Serre, T., Wolf, L., Bileschi, S., Riesenhuber, M., Poggio, T.: Robust object recognition with cortex-like mechanisms. TPAMI (2007) 5 88.
    Google ScholarFindings
  • Shen, L., Yue, Z., Feng, F., Chen, Q., Liu, S., Ma, J.: Msr-net: Low-light image enhancement using deep convolutional network. arXiv (2017) 4 89.
    Google ScholarFindings
  • Simoncelli, E.P., Adelson, E.H.: Noise removal via bayesian wavelet coring. In: ICIP (1996) 3 90.
    Google ScholarLocate open access versionFindings
  • Smith, S.M., Brady, J.M.: SUSANa new approach to low level image processing. IJCV (1997) 3 91.
    Google ScholarFindings
  • Sun, K., Xiao, B., Liu, D., Wang, J.: Deep high-resolution representation learning for human pose estimation. In: CVPR (2019) 5 92.
    Google ScholarFindings
  • Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., Rabinovich, A.: Going deeper with convolutions. In: CVPR (2015) 5 93.
    Google ScholarFindings
  • Tai, Y., Yang, J., Liu, X.: Image super-resolution via deep recursive residual network. In: CVPR (2017) 3 94.
    Google ScholarFindings
  • Tai, Y., Yang, J., Liu, X., Xu, C.: Memnet: A persistent memory network for image restoration. In: ICCV (2017) 3 95.
    Google ScholarLocate open access versionFindings
  • Talebi, H., Milanfar, P.: Global image denoising. TIP (2013) 9 96.
    Google ScholarFindings
  • Tao, X., Gao, H., Shen, X., Wang, J., Jia, J.: Scale-recurrent network for deep image deblurring. In: CVPR (2018) 2 97.
    Google ScholarFindings
  • Tomasi, C., Manduchi, R.: Bilateral filtering for gray and color images. In: ICCV (1998) 3 98.
    Google ScholarFindings
  • Tong, T., Li, G., Liu, X., Gao, Q.: Image super-resolution using dense skip connections. In: ICCV (2017) 4 99.
    Google ScholarFindings
  • Wang, R., Zhang, Q., Fu, C.W., Shen, X., Zheng, W.S., Jia, J.: Underexposed photo enhancement using deep illumination estimation. In: CVPR (2019) 8, 12
    Google ScholarLocate open access versionFindings
  • 100. Wang, S., Zheng, J., Hu, H.M., Li, B.: Naturalness preserved enhancement algorithm for non-uniform illumination images. TIP (2013) 12
    Google ScholarFindings
  • 101. Wang, W., Wei, C., Yang, W., Liu, J.: Gladnet: Low-light enhancement network with global awareness. In: FG (2018) 12
    Google ScholarFindings
  • 102. Wang, X., Girshick, R., Gupta, A., He, K.: Non-local neural networks. In: CVPR (2018) 6
    Google ScholarFindings
  • 103. Wang, X., Yu, K., Wu, S., Gu, J., Liu, Y., Dong, C., Qiao, Y., Change Loy, C.: ESRGAN: enhanced super-resolution generative adversarial networks. In: ECCVW (2018) 4
    Google ScholarFindings
  • 104. Wang, Z., Liu, D., Yang, J., Han, W., Huang, T.: Deep networks for image superresolution with sparse prior. In: ICCV (2015) 4
    Google ScholarLocate open access versionFindings
  • 105. Wang, Z., Chen, J., Hoi, S.C.: Deep learning for image super-resolution: A survey. TPAMI (2019) 3, 10
    Google ScholarLocate open access versionFindings
  • 106. Wei, C., Wang, W., Yang, W., Liu, J.: Deep retinex decomposition for low-light enhancement. BMVC (2018) 4, 8, 12, 13
    Google ScholarFindings
  • 107. Woo, S., Park, J., Lee, J.Y., So Kweon, I.: CBAM: Convolutional block attention module. In: ECCV (2018) 6
    Google ScholarFindings
  • 108. Xiao, B., Wu, H., Wei, Y.: Simple baselines for human pose estimation and tracking. In: ECCV (2018) 5
    Google ScholarLocate open access versionFindings
  • 109. Xiong, Z., Sun, X., Wu, F.: Robust web image/video super-resolution. TIP (2010) 3
    Google ScholarFindings
  • 110. Xu, J., Zhang, L., Zhang, D.: A trilateral weighted sparse coding scheme for real-world image denoising. In: ECCV (2018) 9
    Google ScholarFindings
  • 111. Xu, J., Zhang, L., Zhang, D., Feng, X.: Multi-channel weighted nuclear norm minimization for real color image denoising. In: ICCV (2017) 9
    Google ScholarFindings
  • 112. Yang, J., Wright, J., Huang, T., Ma, Y.: Image super-resolution as sparse representation of raw image patches. In: CVPR (2008) 3
    Google ScholarLocate open access versionFindings
  • 113. Yang, J., Wright, J., Huang, T.S., Ma, Y.: Image super-resolution via sparse representation. TIP (2010) 3
    Google ScholarLocate open access versionFindings
  • 114. Yaroslavsky, L.P.: Local adaptive image restoration and enhancement with the use of DFT and DCT in a running window. In: Wavelet Applications in Signal and Image Processing IV (1996) 3
    Google ScholarFindings
  • 115. Ying, Z., Li, G., Gao, W.: A bio-inspired multi-exposure fusion framework for low-light image enhancement. arXiv preprint arXiv:1711.00591 (2017) 12, 13
    Findings
  • 116. Ying, Z., Li, G., Ren, Y., Wang, R., Wang, W.: A new image contrast enhancement algorithm using exposure fusion framework. In: CAIP (2017) 12
    Google ScholarFindings
  • 117. Yue, Z., Yong, H., Zhao, Q., Meng, D., Zhang, L.: Variational denoising network: Toward blind noise modeling and removal. In: NeurIPS (2019) 9, 10
    Google ScholarLocate open access versionFindings
  • 118. Zhang, K., Zuo, W., Chen, Y., Meng, D., Zhang, L.: Beyond a gaussian denoiser: Residual learning of deep cnn for image denoising. TIP (2017) 2, 3, 9
    Google ScholarLocate open access versionFindings
  • 119. Zhang, K., Zuo, W., Zhang, L.: FFDNet: Toward a fast and flexible solution for CNN-based image denoising. TIP (2018) 3, 9
    Google ScholarLocate open access versionFindings
  • 120. Zhang, L., Wu, X.: An edge-guided image interpolation algorithm via directional filtering and data fusion. TIP (2006) 3
    Google ScholarFindings
  • 121. Zhang, R.: Making convolutional networks shift-invariant again. In: ICML (2019) 7
    Google ScholarFindings
  • 122. Zhang, Y., Zhang, J., Guo, X.: Kindling the darkness: A practical low-light image enhancer. In: MM (2019) 2, 4, 12, 13
    Google ScholarFindings
  • 123. Zhang, Y., Li, K., Li, K., Wang, L., Zhong, B., Fu, Y.: Image super-resolution using very deep residual channel attention networks. In: ECCV (2018) 4, 6, 10, 11, 12
    Google ScholarFindings
  • 124. Zhang, Y., Li, K., Li, K., Zhong, B., Fu, Y.: Residual non-local attention networks for image restoration. In: ICLR (2019) 4, 6
    Google ScholarLocate open access versionFindings
  • 125. Zhang, Y., Tian, Y., Kong, Y., Zhong, B., Fu, Y.: Residual dense network for image restoration. TPAMI (2020) 2, 4
    Google ScholarFindings
Your rating :
0

 

Tags
Comments