Super-resolution through neighbor embedding

By Hong Chang, Dit-Yan Yeung, Yimin Xiong. Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004., 2004

2,173 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/5fbfac24409ce3cb1dbafd0a8227fd1d569ba208

How researchers see this paper

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

Citations in the corpus

Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution (2017, methodology): “Numerous SR methods learn the LR-HR mapping with image pairs collected from external databases using supervised learning algorithms, such as nearest neighbor [10], manifold embedding [2, 5], kernel ridge regression [19], and sparse representation [37, 38, 39].”
Dimensionality Reduction: A Comparative Review (2008, background): “, superresolution [24] and sound source localization [37].”
Image Super Resolution Based on Fusing Multiple Convolution Neural Networks (2017, methodology): “The typical methods include neighbor embedding [7], Markov network [22], kernel regression [4], random forest [26], sparse coding [25][9][10][12], and convolution neural networks.”
Fast Direct Super-Resolution by Simple Functions (2013, methodology): “Notwithstanding much demonstrated success of the algorithms in the literature, existing methods require computationally expensive processes in either searching exemplars [4, 1, 5] or extracting complex features [12, 16, 17, 6].”
Ultra-Resolving Face Images by Discriminative Generative Networks (2016, methodology): “Generic super-resolution: In general, generic single image super-resolution methods have three types: interpolation based methods, image statistics based methods [20,21] and example (patch)-based methods [22,23,24,7,25,26].”
Deep Learning for Single Image Super-Resolution: A Brief Review (2018, methodology): “Neighbor embedding methods [17] proposed by Chang et al.”
Image Super-Resolution With Sparse Neighbor Embedding (2012, methodology): “For the methods in [10] and [13], only the neighborhood size is determined.”
A survey on super-resolution imaging (2011, methodology): “Recently, learning-based techniques are proposed to tackle the SR problem [67–70].”
Conditioned Regression Models for Non-blind Single Image Super-Resolution (2015, methodology): “Neighborhood embedding approaches [2, 3] go without learned dictionaries and directly use all the training data for coding.”
End-to-End Image Super-Resolution via Deep and Shallow Convolutional Networks (2016, methodology): “Based on the neighbor embedding algorithm, works of [8] and [32] super-resolve LR images with the assumption that LR and HR patches lie on low-dimensional nonlinear man-”
Single image super-resolution from transformed self-exemplars (2015, methodology): “Existing methods have employed various learning algorithms for learning this LR to HR mapping, including nearest neighbor approaches [14], manifold learning [6], dictionary learning [41], locally linear regression [38] [33] [34], and convolutional networks [9].”
Multi-scale dictionary for single image super-resolution (2012, background): “• Example-based SR approaches (e.g., [7] [3] [17] [10]) assume that the high-frequency details lost in the LR image can be effectively predicted from a set of LR and HR training image pairs.”
Shepard Convolutional Neural Networks (2015, unknown): “Methods compared: Bicubic, K-SVD [13], NE+NNLS [14], NE+LLE [15], ANR [16], A+ [11], SRCNN [4], Our ShCNN”
Learning Enriched Features for Fast Image Restoration and Enhancement (2022, background): “Prior to the deep-learning era, numerous super-resolution (SR) algorithms have been proposed based on the sampling theory [45], [46], edge-guided interpolation [47], [48], natu-ral image priors [49], [50], patch-exemplars [51], [52] and sparse representations [53], [54].”
Bidirectional Recurrent Convolutional Networks for Multi-Frame Super-Resolution (2015, methodology): “• NE+LLE: locally linear neighbor embedding [7].”
Single-Image Super-Resolution Using Sparse Regression and Natural Image Prior (2010, background): “Ç”
DeepSD: Generating High Resolution Climate Change Projections through Single Image Super-Resolution (2017, methodology): “mages to produce exemplar patches. „is approach has evolved into di‡erent variations of nearest neighbor techniques between lowand high-resolution patches through what is known as dictionary learning [7, 11, 43]. Dictionary learning approaches to SR are analogous to those presented by weather classi”
Super resolution using edge prior and single image detail synthesis (2010, unknown): “These approaches have shown great promise and have been applied to SR in various ways, including generic detail synthesis for upsampling [10, 19, 9, 2], edge-focused detail synthesis [24], imposing consistency on synthesized detail [24, 5], and targeting multiple low-resolution images [29].”