Image Super-Resolution With Sparse Neighbor Embedding

By Xinbo Gao, Kaibing Zhang, Dacheng Tao, Xuelong Li. IEEE Transactions on Image Processing, 2012

335 citations recorded by the source corpus.

Until now, neighbor-embedding-based (NE) algorithms for super-resolution (SR) have carried out two independent processes to synthesize high-resolution (HR) image patches. In the first process, neighbor search is performed using the Euclidean distance metric, and in the second process, the optimal weights are determined by solving a constrained least squares problem. However, the separate processes are not optimal. In this paper, we propose a sparse neighbor selection scheme for SR reconstruction. We first predetermine a larger number of neighbors as potential candidates and develop an extended Robust-SL0 algorithm to simultaneously find the neighbors and to solve the reconstruction weights. Recognizing that the k-nearest neighbor (k-NN) for reconstruction should have similar local geometric structures based on clustering, we employ a local statistical feature, namely histograms of oriented gradients (HoG) of low-resolution (LR) image patches, to perform such clustering. By conveying local structural information of HoG in the synthesis stage, the k-NN of each LR input patch is adaptively chosen from their associated subset, which significantly improves the speed of synthesizing the HR image while preserving the quality of reconstruction. Experimental results suggest that the proposed method can achieve competitive SR quality compared with other state-of-the-art baselines.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/5359c2aeddcb1ee60239dbdc4b84a3cc3d75ec2a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Enhanced Deep Residual Networks for Single Image Super-Resolution (2017, methodology): “Those learning methods rely on techniques ranging from neighbor embedding [3, 2, 7, 21] to sparse coding [31, 32, 27, 33].”
Super-resolution: a comprehensive survey (2014, methodology): “Structural Similarity Measures [220], [270], [275], [293], [355], [366], [420], [428], [439], [449], [456], [485], [493], [495], [507], [520], [522], [524], [527], [552], [554], [555], [557], [558], [561], [563], [568], [569], [570], [571], [572], [579], [586], [590], [595], [601], [608], [612]”
Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network (2016, neutral): “Recent popular SISR methods can be classified into edge-based [35], image statisticsbased [9, 18, 46, 12] and patch-based [2, 43, 52, 13, 54, 40, 5] methods.”
Denoising Prior Driven Deep Neural Network for Image Restoration (2018, background): “For example, the IR problem is a denoising problem [1]–[5] when A is an identical matrix and becomes a deblurring problem [6]–[9] when A is a blurring matrix/operator, or a super-resolution problem [8], [10]–[12] when A is a subsampling matrix/operator.”
A Survey of Sparse Representation: Algorithms and Applications (2015, methodology): “[158] proposed a sparse neighbor embedding method, which incorporated the sparse neighbor search and HoG clustering method into the process of image super-resolution reconstruction.”
DDcGAN: A Dual-Discriminator Conditional Generative Adversarial Network for Multi-Resolution Image Fusion (2020, background): “For instance, the Euclidean distance suffers from relatively blurred results by averaging all plausible outputs [30].”
Image Super-Resolution Using Dense Skip Connections (2017, methodology): “In addition to sparsity-based methods, other sophisticated learning techniques have been developed to model the mapping from LR to HR space, including neighbor embedding [4], random forest [20] and convolutional neural network [2].”
Analyzing Perception-Distortion Tradeoff using Enhanced Perceptual Super-resolution Network (2018, methodology): “le dictionary. HR images from the web with similar contents were used with-in a structure-aware matching criterion to super-resolve landmark images in [48]. The class of neighbor embedding approaches [8,3,17,44,45] aim to nd similar looking LR training patches from a low dimensional manifold and the”
Dual Reconstruction Nets for Image Super-Resolution with Gradient Sensitive Loss (2018, methodology): “Besides, there are some other methods, such as sparsity-based techniques [12, 19] and neighborhood embedding [11, 41], which have been widely used in real-world applications.”
Hyperspectral Image Spatial Super-Resolution via 3D Full Convolutional Neural Network (2017, methodology): “In the past decades, the SR of traditional color images has gained great attention and many algorithms have been developed, such as iterative back projection (IBP) based on reconstruction [36,37] and sparse representation based algorithms [38,39].”