Low-Complexity Single-Image Super-Resolution based on Nonnegative Neighbor Embedding

By Marco Bevilacqua, Aline Roumy, Christine M. Guillemot, Marie-Line Alberi-Morel. 2012

2,911 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

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

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network (2016, methodology): “Mean opinion score (MOS) testing In all conducted MOS tests we have asked 26 human raters to assign a score from 1 (Bad) to 5 (Excellent) to reconstructions of the 4× downsampled versions of images from Set5, Set14 and BSD100.”
Perceptual Losses for Real-Time Style Transfer and Super-Resolution (2016, methodology): “We evaluate all models on the standard Set5 [65], Set14 [66], and BSD100 [46] datasets.”
AutoGAN-Distiller: Searching to Compress Generative Adversarial Networks (2020, neutral): “…end for 12: # Second Step: Search 13: for t ← 1 to T 2 do 14: Get a batch of data X 1 from χ 1 15: Get a batch of data X from χ 27: w ) 29: end for SR benchmarks, including Set5 (Bevilacqua et al., 2012), Set14 (Zeyde et al., 2010), BSD100 (Martin et al., 2001) and Urban100 (Huang et al., 2015).”
Hierarchical Neural Architecture Search for Single Image Super-Resolution (2020, methodology): “To show the effectiveness of out method, we conduct comparisons on ﬁve benchmark datasets, including Set5 [36], Set14 [37], BSD100 [38], Urban100 [39], and Manga109 [40].”
Deep Image Prior (2017, methodology): “We evaluate super-resolution ability of our approach using Set5 [2] and Set14 [32] datasets.”
Image Super-Resolution Using Deep Convolutional Networks (2014, methodology): “We use the Set5 [2] as the validation set.”
EnhanceNet: Single Image Super-Resolution Through Automated Texture Synthesis (2016, background): “In recent years, popular approaches include exemplar-based models that either exploit recurrent patches of different scales within a single image [13, 17, 22, 56] or learn mappings between low and high resolution pairs of image patches in external databases [3, 5, 14, 27, 51, 58, 63].”
Accurate Image Super-Resolution Using Very Deep Convolutional Networks (2015, methodology): “Neighbor embedding [4, 15] methods interpolate the patch subspace.”
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks (2018, methodology): “We evaluate our models on widely used benchmark datasets – Set5 [42], Set14 [43], BSD100 [44], Urban100 [45], and the PIRM self-validation dataset that is provided in the PIRM-SR Challenge.”
Enhanced Deep Residual Networks for Single Image Super-Resolution (2017, methodology): “We also compare the performance on four standard benchmark datasets: Set5 [2], Set14 [33], B100 [17], and Urban100 [10].”
Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution (2017, methodology): “We carry out extensive experiments using 5 datasets: SET5 [2], SET14 [39], BSDS100 [1], URBAN100 [15] and MANGA109 [23].”
Image Super-Resolution Using Very Deep Residual Channel Attention Networks (2018, neutral): “For testing, we use five standard benchmark datasets: Set5 [36], Set14 [37], B100 [38], Urban100 [22], and Manga109 [39].”
A Deep Journey into Super-resolution (2019, methodology): “• Set5 [88] is a classical dataset and only contains ﬁve test images of a baby, bird, butterﬂy, head, and a woman.”
Deep Networks for Image Super-Resolution with Sparse Prior (2015, methodology): “Comparison with State of the Arts We compare the proposed CSCN with other recent SR methods on all the images in Set5, Set14 and BSD100 for different upscaling factors.”
Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescaling (2021, neutral): “Downscaling & Upscaling Param Set5 [4] Set14 [48] BSD100 [33] Urban100 [14] DIV2K [1]”
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.”
Unsupervised Degradation Representation Learning for Blind Super-Resolution (2021, methodology): “Following [12], we used 800 training images in DIV2K [1] and 2650 training images in Flickr2K [36] as the training set, and included four benchmark datasets (Set5 [4], Set14 [39], B100 [27] and Urban100 [19]) for evaluation.”