Académique Documents
Professionnel Documents
Culture Documents
Signal Processing
journal homepage: www.elsevier.com/locate/sigpro
a r t i c l e in f o abstract
Article history: This paper proposes a new approach to single-image super-resolution (SR) based on
Received 4 June 2013 generalized adaptive joint kernel regression (G-AJKR) and adaptive dictionary learning.
Received in revised form The joint regression prior aims to regularize the ill-posed reconstruction problem by
30 September 2013
exploiting local structural regularity and nonlocal self-similarity of images. It is composed
Accepted 18 November 2013
of multiple locally generalized kernel regressors defined over similar patches found in the
Available online 27 December 2013
nonlocal range which are combined, thus simultaneously exploiting both image statistics
Keywords: in a natural manner. Each regression group is then weighted by a regional redundancy
Single-image super-resolution measure we propose to control their relative effects of regularization adaptively. This joint
Face hallucination
regression prior is further generalized to the range of multi-scales and rotations. For
Face recognition
robustness, adaptive dictionary learning and dictionary-based sparsity prior are intro-
Joint kernel regression
Dictionary learning duced to interact with this prior. We apply the proposed method to both general natural
images and human face images (face hallucination), and for the latter we incorporate a
new global face prior into SR reconstruction while preserving face discriminativity. In both
cases, our method outperforms other related state-of-the-art methods qualitatively and
quantitatively. Besides, our face hallucination method also outperforms the others when
applied to face recognition applications.
& 2013 Elsevier B.V. All rights reserved.
0165-1684/$ - see front matter & 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.sigpro.2013.11.042
C. Huang et al. / Signal Processing 103 (2014) 142154 143
Generally, pYjX is modeled by the Gaussian distribu- discards the further potential enabled by the higher-order
tion, thus maximizing pYjX boils down to minimizing the statistics. Besides, it needs a separate deblurring process
data constraint [2] Y DHX22 . On the other hand, p(X) which is ill-posed by itself. In our previous work [18],
codes the prior knowledge that we want to impose in the we proposed an Adaptive Joint Kernel Regression (AJKR)
HR space. Typically, the task of SR reconstruction is method, combining a set of coherent NLM-generalized
formulated as a regularized least-square optimization local regressors in the nonlocal range with higher-order
problem as follows: information (i.e. regional redundancy) injected in. By further
integrating adaptive dictionary learning under the MAP
^ arg minY DHX2 CX;
X 3
X
2 framework, this algorithm produces superior results than
NLKR and excludes the necessity of separate deblurring.
where is the parameter balancing the effects of the data However, it only builds kernel regressors at the same scale
constraint and the regularization term C(X). Most of the and rotation. To exploit the full potential offered by such
past works focus on designing different formulations joint regression, the core algorithm should be generalized.
of C to regularize the ill-posed reconstruction problem. For face hallucination, the above methods dealing with
Currently, the single-image SR methods can be mainly general natural images cannot be readily applied due to
categorized into three classes: interpolation-based meth- the ignorance of the special properties of face images. This
ods, reconstruction-based methods, and example-based problem was first addressed in the pioneering work of
methods. Interpolation techniques (e.g. [6]) are simple and Baker and Kanade [1]. They adopted an image pyramid to
fast but tend to blur the fine details. The reconstruction- learn a prior on the gradient distribution of frontal face
based methods (e.g. [710]) follow the form of Eq. (3) and images using the Bayesian theory. However, the HR image
how to design a good image prior is always an essential prediction is pixelwise which causes discontinuities and
issue; C is usually a smoothness constraint. The example- artifacts. To generate high-quality HR face images, the
based methods (e.g. [9,1115]) hallucinate detailed textures current face hallucination methods usually involve two
from a training set of LR/HR image or patch pairs. However, steps. The first step reconstructs global faces in the face
such methods strongly rely on the chosen dataset for subspace using MAP criterion [11,2] or manifold learning
satisfactory results. methods. Principal Component Analysis (PCA) [2,3] is
Many example-based methods directly or implicity use widely used for face modeling. Classic manifold learning
a co-occurrence prior to constrain the correspondence methods include Locality Preserving Projections (LPPs) [5],
between LR and HR patches. For example, Yang et al. [11] Canonical Correlation Analysis (CCA) [4] and so on. While
explored the sparse representation of LR patches over an neighborhood preservation and correlation maximization
LR dictionary, and used the same representation coeffi- are the only concerns, discriminativity is often lost and the
cients to generate the HR output, but the result usually frequently used PCA, for example, yields results like mean
suffers from inconsistency between neighboring patches. face; the second step produces a residue image to recover
Other natural image priors are also studied in the litera- details [2,4,5,11].
ture. The gradient profile prior is developed in [8] to This paper focuses on SR from a given LR version of
preserve sharp edges, but is limited in modeling the visual general natural image or face image. Similar to our pre-
complexity of real images. Later, priors of image self- vious AJKR method [18], we address this problem from the
similarities and local/nonlocal regularities have been viewpoints of learning good regression priors and robust
exploited for more robust estimation. In [9], the nonlocal dictionaries. However, we generalize AJKR in two ways: (1)
self-similarity properties both within and across spatial to extend the regression range to multi-scales and rota-
scales are fully exploited, but the local regularities are tions, obtaining a Generalized AJKR (G-AJKR) method, and
neglected. Zhang et al. [10] improved by assembling the (2) to incorporate a new global structure prior of human
Steering Kernel Regression [16] (SKR)-based local prior faces into the G-AJKR method for face hallucination, while
and Nonlocal Means [17] (NLM)-based nonlocal prior, preserving individual discriminativity based on Partial
whose connection, however, remains loose. Least Squares (PLS) [19], which is very important when
Another trend in SR is to combine the reconstruction- applied to face recognition applications.
and example-based methods (usually dictionary induced) The remainder of the paper is organized as follows.
into a unified framework to produce more compelling Section 2 reviews related works on dictionary and mani-
results. In fact, SR can be viewed as a regression problem fold learning as the development that follows relies on
aiming to map LR images to target HR images. Then in this them. Section 3 details our G-AJKR framework and its
sense, dictionary-based methods do local regression using extension to face hallucination. Experimental results of SR
bases learned from an external database or the input image on generic and face images with applications in face
itself, while regression models directly estimate HR pixels recognition are provided in Section 4. We conclude the
(kernel learning) or regularize the estimator. As for the paper in Section 5.
regression models, examples include SKR [16], Gaussian
Process Regression (GPR) [12], Kernel Ridge Regression 2. Related works
(KRR) [13] and Non-Local Kernel Regression (NLKR) [14],
and they can all be effectively exploited as a prior for SR 2.1. Dictionary learning
reconstruction. Among them, NLKR overcomes the draw-
backs of the literature [10] by unifying the local and nonlocal Learning a good dictionary is important for example-
priors into a single model in a complimentary way, but it based methods to do local regression using the learned
144 C. Huang et al. / Signal Processing 103 (2014) 142154
Fig. 1. Two dimensional embeddings of PCA coefficients of LR (first row) and HR (second row) face images by different subspace methods.
bases. Traditional choices are the analytically designed effort on neighborhood preservation or correlation maximiza-
wavelet bases which lack sufficient flexibility to a given tion may congregate different neighbors together in the
image. Other methods learn a dictionary (usually over- projected subspace, making the local topology recovery
complete) from an image database using techniques like therein very difficult. Taking CCA for example, imagine an
K-SVD [20] and PCA [21], but the flexibility is still limited extreme case when the first coordinates of X and Y are
since the dictionary is only learned to perform well on perfectly correlated with the others almost uncorrelated, CCA
average. Online dictionary learning from the given image will give the first coordinate as the principle direction which
itself offers a promising alternative to exploit rich informa- projects all the data points in X and Y to a common single
tion contained in the input [22,23]. One drawback is that point, making it impossible to recover the neighborhood
the learning process easily runs into the risk of building structure. This will result in either wrong neighbor selection
dictionaries with many artifacts under image corruptions. while recovery in the subspace YV with an unfaithful face, or
indistinguishable reconstruction results in the subspace XU
2.2. Manifold learning for face hallucination with a near-mean face, both hampering face recognition with
reduced discriminativity. Recently, Partial Least Squares (PLS)
The most popular modeling method in face hallucina- [19] was proposed and successfully applied to face recognition
tion is PCA [2,3], but it is holistic and tends to yield faces and multi-modal recognition [26]. It finds normalized bases U
like the mean. Since face images are shown to reside on and V to maximize the covariance
a low-dimensional nonlinear manifold, researchers are
arg max covXU; YV arg max varXU corrXU; YV2 varYV;
inspired to use manifold learning to hallucinate global fU;Vg fU;Vg
faces. Typical methods include Locally linear embedding
s:t: J U J J V J 1: 5
(LLE) [24,25] and LPP [5] projecting onto a subspace that
preserves neighborhood relationships. A major assump- Clearly, PLS tries to maintain correspondence as well as
tion of applying all these methods to infer HR faces is that preserve the variance. Fig. 1 shows LR and HR subspaces
LR and HR manifolds have similar local topologies so that obtained from the PCA coefficients of real face images. As can
the HR image can be represented as a linear combination be seen, the correlation (local neighborhood similarity
of neighbors using the same weights derived in the LR assumption) does not hold well for LPP and LLE, also with
space. To strengthen this assumption in practice, CCA [4] little discrimination preserved. Although the correlation is
finds an optimal subspace that maximizes the correlation perfectly maintained by CCA, the projections still congregate.
between LR and HR images. Specifically, CCA finds two On the other hand, PLS preserves both correlation and
bases U and V to linearly map the two sets of vectors X and discriminativity very well.
Y to a common subspace where the correlation is max-
imized
3. Proposed G-AJKR framework for single-image SR
covXU; YV
arg max corrXU; YV2 arg fU; Vg ; 4
fU;Vg max varXU varYV
To make better use of image local structural regularity and
where cov; is the covariance operator. nonlocal self-similarity, a G-AJKR method is proposed in
Unfortunately, CCA and the other subspace methods this section. An overview of the proposed method is shown
mentioned above suffer from a common drawback: too much in Fig. 2. A new joint regression prior is learned across scales
C. Huang et al. / Signal Processing 103 (2014) 142154 145
Fig. 2. (a) Graphical illustration of the G-AJKR framework, where the reference patch is marked as R; (b) block diagram of our generic image SR
algorithm; (c) extension to face hallucination by introducing a new global face structure prior based upon PLS.
and rotations and is weighted by the regional redundancy 3.1. Generic image SR
measure (Fig. 2(a)), and the adaptive dictionary learning is
integrated (Fig. 2(b)). We also study how to introduce a new 3.1.1. Review of the previous AJKR method
global face structure prior into G-AJKR for it to be tailored The AJKR method in our previous work [18] combines
towards face hallucination (Fig. 2(c)). cues from local and nonlocal image priors, which are
146 C. Huang et al. / Signal Processing 103 (2014) 142154
inspired by SKR [16] and NLM [17], respectively. It enables matrix K, we can obtain the matrix form:
more reliable and robust results than similar unifying ^ arg minY DHX2 I KX2 ;
X 2 2 10
methods by simultaneously exploiting both image priors X
in a higher-order collaborative manner. Let xi denote the where I is the identity matrix.
location of the ith pixel in the HR grid, Yi denotes the pixel Considering the degree of patch redundancy for the
observation at xi, and Y i is the patch vector of pixels in xi0 s joint regression in Eq. (9) varies significantly across differ-
local neighborhood N xi . Then the joint kernel regression ent regions within an image, we further propose an
model is formulated as explicit measure of regional redundancy to determine
the confidence of each regression group that gives
a^ i arg min wN 2
ij Y j aWN ; 6 !
a
Xi Xj 2WG
j
j A Pxi
N 2 N
Ri wij ; wij exp 2
; 11
where Yj is the similar patch to the one at xi (including j A Pxi hn
itself) found in a nonlocal range Pxi within the same which can also be regarded as penalizing the patch
N
image scale. It performs regression on each similar patch distances due to the way wij are calculated. Obviously,
using the polynomial bases (say second-order) from the smaller the distances are (the larger Ri), the more
Taylor expansion with regression coefficients a, and com- similar the grouped patches are and the more patch
N
bines them using the patch similarity weights wij defined redundancy there is in the nonlocal region. Usually, edges
by and smooth areas have large values while textures have
! small values. We then use it for adaptive regularization:
Y i Y j 2WG
wN
ij exp ; 7 ^ arg minY DHX2 I KX2 ;
X 12
2 2 R
hn X
where e1 is a vector with the first element equal to one Patch rotation is simply achieved by bicubic interpola-
and the rest zero. tion. To search at multi-scales, we generate a multi-
T
Defining the row vectors kij eT1 T j A Pxi wN N
ij W j
1 resolution image hierarchy of decreasing resolutions {Xs}
N
T wN
ij W j to be the equivalent kernels with which we scaled down by operator Ds with scale factors at 1.25-s,
perform the regression for xi, we can plug them into the s A 0; S. Here, Ds is a patch downsampling operator which
T
SR optimization function in Eq. (3) to act as the regulariza- keeps the patch center on the LR grid, while Ds is a patch
tion C : upsampling operator with zero-padding [14]. Then for the
reference patch centered at xi on the current image plane,
n s
^ arg minY DHX2 X i kT Xj 2 ; we compare to find its nonlocal neighbors Xj of the same
X 2 ij 2 9 s
X i1 j A Pxi patch size J at location 1.25 xj on the image plane of Xs.
This leads to a generalized nonlocally similar patch set
where Xi denotes the pixel to be estimated at location xi, and P s xi , and further to P s; xi when the rotated patch
Xj is its similar patch centered at xj found in the nonlocal versions Xs; j are considered (see Fig. 2(a) for example).
T
range. By properly arranging kij to the equivalent kernel This way, we obtain a generalized joint kernel regression
C. Huang et al. / Signal Processing 103 (2014) 142154 147
Fig. 3. (a) Centroids of the offline PCA dictionary B0; (b) examples of the online PCA dictionaries B1, where the 8 first atoms are shown; (c) PSNR curves
( 3) versus iterations for different dictionary learning schemes on the lena image corrupted with a Gaussian blur (sb 2) and Gaussian noise (sn 5).
148 C. Huang et al. / Signal Processing 103 (2014) 142154
tolerable to noise or blur that may cause e.g. grouping intermediate subspace
errors. M
ch i0 cH
i0 : 22
0
3.2. Face hallucination i 1
Subsequently, we can reconstruct the PCA coefficients
The face images differ from the general natural images bh of the HR image and the hallucinated HR global face
image X^ as
in that the face images are more regular in structure,
therefore introducing a global facial structure prior can be bh UUT 1 Uch ; ^ H EH bh :
X 23
conductive. Following [2,4,5,11], we also propose a two-
step face hallucination method, where the first step con- To the best of our knowledge, it is the first time that PLS
strains and reconstructs the global face in a discriminativ- is used for the face hallucination problem. The PLS-based
ity preserving subspace. The learned subspace can face reconstruction instead of simple bicubic interpolation
introduce a strong facial prior to p(X) in Eq. (2) describing is implemented as the first step, as shown in Fig. 2(c), to
the main characteristics of face images. incorporate a good global face prior. Then the proposed
As mentioned in Section 2.2, the commonly used mani- G-AJKR algorithm follows to recover details. From the
fold learning methods LLE and LPP potentially suffer from a Bayesian viewpoint in Eq. (2), G-AJKR further imposes
great loss in the discriminative ability. Since CCA only local/nonlocal priors to p(X) and applies the reconstruc-
correlates the LR and HR subspaces, it can also fail to tion constraint for pYjX using Eq. (17).
differentiate between the subspace projections. As a result,
the hallucinated facial features by these subspace methods 4. Experimental results
may not be faithful to the ground truth but are more like the
mean features or unexpected ones. We here propose to use In this section we evaluate the effectiveness of our
PLS [19] to learn an intermediate subspace, which strikes a method G-AJKR for single-image SR. We first give an
balance between the objectives of CCA and PCA by main- illustrative example to demonstrate the superiority of the
taining correlation while capturing the projection variations generalization to the range of multi-scales and rotations.
(discrimination). By using the co-occurrence assumption, Then we compare with several related as well as state-of-
neighbor-based reconstruction in such subspace will gener- the-art algorithms, on several standard test images and
ate faithful (or unique) visual features, which are crucial for real-world images, both qualitatively and quantitatively.
the task of face recognition. We first learn PCA models for For color images all the test algorithms are applied to the
the LR/HR training images, with the corresponding mean illuminance channel only, where the quantitative results
faces L and H , and orthogonal eigenvectors EL and EH. Thus are carried out in terms of the metrics of PSNR and
LR/HR faces are represented as linear combinations of Structural SIMilarity index (SSIM) [29]. Since in real-
eigenfaces using coefficients bL and bH world SR tasks the observed LR images are often contami-
L nated by noise, the robustness of SR methods with respect
Y L EL b ;
to noise is also evaluated. Finally, the face hallucination
H H H
X E b : 18 and recognition performance are presented.
Suppose we have collected from training set the PCA In all experiments we use HR patches of 7 7 pixels
L Q H Q (L49) with a 4-pixel overlap for both local kernel
coefficients {bi }i 1 and {bi }i 1, then PLS is applied to them
to find two normalized bases V and U to maximize the regression and patch matching. We set the support of
L the nonlocal searching to be 15-nearest neighbors in a
covariance using Eq. (5). Projecting the PCA coefficients ci
H
and ci by these bases into a common subspace, we have window of size 21 21 across all pyramid levels. Since the
high dimensionality of the 4D search space imposes a large
L H
cLi VT bi ; cH T
i U bi ; 19 computational burden, we only set S3 (i.e. 4 hierarchy
where
L
ci
and
H
ci
are the subspace projections whose levels) and rotation range A f 901; 451; 01; 451; 901g
correlation is maximized with the discriminativity pre- without much performance degradation. Besides, we
served. Fig. 2(c) illustrates this relationship which is bene- apply the 4D search only to patches with high intensity
ficial to the neighbor-based reconstruction. Given an input variance (threshold 16). This is due to that the SR effect is
LR face image Y with its PCA coefficients bl computed, we expressed mostly in highly detailed image regions (e.g. edges
obtain its PLS projection cl VT bl . Then for cl we seek its M and textures) rather than the uniform and low-frequency
M ones (we apply 2D search for them). As a result, it typically
nearest neighbors fcLi0 gi0 1 in the trained subspace and
corresponding weights f i0 gM takes about 4 min to process a 256 256 image on a PC
i0 1
2 (3.2 GHz, Pentium IV) with our unoptimized MATLAB code.
Other parameters are set as 40, 0.25, hn 15. The
M
M
arg mincl i0 cLi0 ; s:t: i0 1: 20
fg 0 0 synthetic LR images were generated from original images
i 1 i 1 2
by a truncated 7 7 Gaussian kernel (sb 1.6) and down-
The closed form solution of the weight is given by [25] sampled by a factor of 3. A Gaussian noise (sn 5) was also
j0 Ai0 j01 added. For solving the optimization problem in Eq. (17), we
i0 1
; Ai0 j0 cl cLi0 cl cLj0 : 21 use the iterative shrinkage algorithm [28].
lm Alm
Our face hallucination experiments were performed on
M
Using the same weights for the corresponding fcH g
i0 i0 1
, frontal view face images from the CAS-PEAL [30] database
we can reconstruct the HR projection features in the and FERET [31] database. For the CAS-PEAL database, we
C. Huang et al. / Signal Processing 103 (2014) 142154 149
randomly selected Q500 images (one per person) with 4.1. Generic image SR experiments
normal expression to train the PLS subspace, and other 40
images (disjoint from the training set) for testing; for the Fig. 4 validates the efficacy of our 4D generalization of
FERET database, Q800 images were randomly selected the basic algorithm on noisy Parrot image. We compare the
for subspace training with other 403 images for testing. All results of AJKR and G-AJKR using different scales and
the face images were aligned by the eye positions and rotations. It is shown that the multi-scale-and-rotation
mouth center and cropped to 128 96 pixels (HR) and version of AJKR preserves sharper edges and more faithful
32 24 pixels (LR). Then the training set for PLS learning is details than the original AJKR, with the G-AJKR only across
composed of LR/HR image pairs, and we zoom the 32 24 rotations being in-between.
LR test image by 4 times. To prepare the offline dictionary Next we compare our method against four regression-
B0 for our G-AJKR algorithm in this scenario, we sampled based methods, GPR [12], KRR [13], NLKR [14] and Zhang
patches from the HR training images and learned the et al. [10]0 s method, two dictionary-based methods, Cen-
dictionary following [21]. In the global face reconstruction tralized Sparse Representation (CSR) [22] and Sparse Coding
phase, we use PCA which retains 98% of the variance and (SC) method [11], and three state-of-the-art methods, Shan
200 PLS bases are kept for the subsequent projection. The et al. [7], Glasner et al. [9] and Freedman and Fattal [15].
neighborhood size is set as M50. Among them, the GPR and KRR are two recent regression
Fig. 4. Comparison of SR results ( 3) on noisy Parrot image obtained by AJKR [18] and G-AJKR using different scales and rotations. (a) LR input; (b) AJKR
(PSNR: 29.62 dB, SSIM: 0.892); (c) G-AJKR across rotations but at the same scale (S 0) (PSNR: 29.67 dB, SSIM: 0.896); (d) G-AJKR across rotations and
scales (PSNR: 29.78 dB, SSIM: 0.903).
Table 1
Comparison of SR results ( 3, PSNR/SSIM) for the noiseless case.
Table 2
Comparison of SR results ( 3, PSNR/SSIM) for the noisy case.
methods that capture from input image the mapping authors or downloaded from their websites, and we used
between LR and HR patches via Gaussian process regression their default parameter setups.
and sparse kernel regression, respectively. All the compared Table 1 and 2 show the quantitative results for the noise-
results were reproduced by the codes provided by the less and noisy cases, respectively. Since the implementation of
Fig. 5. Visual comparison of SR results ( 3). (a) Noiseless case; (b) noisy case. The PSNR and SSIM results of all methods compared can be seen accordingly
in Tables 1 and 2.
Fig. 7. Visual comparison of hallucinated global faces by different subspace methods on the CAS-PEAL database. (a) LR input; (b) reconstruction result
directly in the PCA space without projection onto other subspaces; (c) PCA LPP; (d) PCALLE; (e) PCA CCA; (f) PCAPLS; (g) original HR image.
Fig. 8. Visual comparison between our two-step face hallucination algorithm with only the G-AJKR algorithm on the CAS-PEAL database. (a) LR input; (b)
global hallucination result; (c) result using the two-step algorithm; (d) result using the G-AJKR algorithm only; (e) original HR image.
152 C. Huang et al. / Signal Processing 103 (2014) 142154
NLKR is not available, we do not include this method in our the state-of-the-art methods of Freedman and Fattal, Shan
quantitative comparisons. Visual quality comparison with its et al. and Glasner et al. our improvements are also evident.
reported result on a real-world image will be shown next. As Fig. 5(b) shows the noise robustness of our method while
can be seen from the tables, our method constantly outper- preserving edges and small details. Again, this can be
forms the others across all metrics for both cases, with the attributed to the full and adaptive exploitation of image
largest PSNR improvements (over GPR) at 4.09 dB (noiseless) self-similarity and the responsive dictionary scheme. More
and 2.89 dB (noisy) on average which are quite significant. For challenging results ( 4 magnifications) on real LR images
the noisy input, not only must image details be recovered, but are shown in Fig. 6. We can see that our method outper-
also the noise need to be suppressed. Methods like SC tend to forms all the others in terms of visual plausibility.
magnify noise during image up-sampling, which lead to lower
results than ours. 4.2. Face hallucination and recognition performance
Fig. 5 visually illustrates the performance difference of
our method to the others. In Fig. 5(a), our method Face hallucination can handle more challenging tasks
synthesizes more visual details and sharpen edges without than generic image SR due to the regular face structure.
blur compared with the three regression methods GPR, Our two-step method first hallucinates a global face in the
KRR and Zhang et al.0 s method. This is due to our coherent learned face subspace which incorporates the special
and collaborative use of regression priors and the rich properties of faces and compensates for the lost informa-
redundancies found across scales and rotations. Our result tion in the input. Fig. 7 compares different subspace
is also free of jaggy and ringing artifacts with the SC and methods for global face hallucination, whose subspace
CSR methods respectively, which validates the benefits of dimension and neighborhood size are the same with ours.
our adaptive dictionary learning scheme (it usually Clearly, after projecting PCA coefficients onto subspaces
improves the baseline performance of G-AJKR by about learned by LPP, LLE and CCA the resultant faces do not look
0.15 dB as in our previous AJKR case [18]). In comparison to like the original face but like a fused one, which means the
Fig. 9. Comparison of hallucination results on the CAS-PEAL (rows 12) and FERET (rows 34) databases along with the corresponding average PSNR
values. (a) LR input; (b) Wang and Tang (average PSNR: 25.33 dB); (c) Zhuang et al. (average PSNR: 29.27 dB); (d) NE (average PSNR: 30.32 dB); (e) SC
(average PSNR: 30.94 dB); (f) our method (average PSNR: 32.84 dB); (g) original HR image.
C. Huang et al. / Signal Processing 103 (2014) 142154 153
[2] C. Liu, H.Y. Shum, W.T. Freeman, Face hallucination: theory and [18] C. Huang, X. Ding, C. Fang, Single-image super-resolution via
practice, Int. J. Comput. Vis. 75 (2007) 115134. adaptive joint kernel regression, in: Proceedings of the British
[3] X. Wang, X. Tang, Hallucinating face by eigentransformation, IEEE Machine Vision Conference, 2013.
Trans. Syst. Man Cybern. Part C 35 (3) (2005) 425434. [19] R. Rosipal, N. Krmer, Overview and recent advances in partial least
[4] H. Huang, H. He, X. Fan, J. Zhang, Super-resolution of human face squares, in: Proceedings of International Conference on Subspace,
image using canonical correlation analysis, Pattern Recognit. 43 (7) Latent Structure and Feature Selection, 2006, pp. 3451.
(2010) 25322543. [20] J. Mairal, M. Elad, G. Sapiro, Sparse representation for color image
[5] Y. Zhuang, J. Zhang, F. Wu, Hallucinating faces: LPH super-resolution restoration, IEEE Trans. Image Process. 17 (1) (2008) 5369.
and neighbor reconstruction for residue compensation, Pattern [21] W. Dong, L. Zhang, G. Shi, X. Wu, Image deblurring and super-
Recognit. 40 (11) (2007) 31783194. resolution by adaptive sparse domain selection and adaptive reg-
[6] X. Li, M.T. Orchard, New edge-directed interpolation, IEEE Trans. ularization, IEEE Trans. Image Process. 20 (7) (2011) 18381857.
Image Process. 10 (10) (2001) 15211527. [22] W. Dong, L. Zhang, G. Shi, Centralized sparse representation for
[7] Q. Shan, Z. Li, J. Jia, C.K. Tang, Fast image/video upsampling, ACM image restoration, in: Proceedings of IEEE International Conference
Trans. Graph. 27 (5) (2008) 17. on Computer Vision, 2011, pp. 12591266.
[8] J. Sun, Z. Xu, H. Shum, Image super-resolution using gradient profile [23] P. Chatterjee, P. Milanfar, Clustering-based denoising with locally
prior, in: Proceedings of IEEE Conference on Computer Vision and learned dictionaries, IEEE Trans. Image Process. 18 (7) (2009)
Pattern Recognition, 2008, pp. 18. 14381451.
[9] D. Glasner, S. Bagon, M. Irani, Super-resolution from a single image, [24] H. Chang, D.Y. Yeung, Y. Xiong, Super-resolution through neighbor
in: Proceedings of IEEE International Conference on Computer
embedding, in: Proceedings of IEEE Conference on Computer Vision
Vision, 2009, pp. 349356.
and Pattern Recognition, 2004, pp. 275282.
[10] K. Zhang, X. Gao, D. Tao, X. Li, Single image super-resolution with
[25] L.K. Saul, S.T. Roweis, Think globally, fit locally: unsupervised
non-local means and steering kernel regression, IEEE Trans. Image
learning of low dimensional manifolds, J. Mach. Learn. Res. 4 (2003)
Process. 21 (11) (2012) 45444556.
119155.
[11] J. Yang, J. Wright, T.S. Huang, Y. Ma, Image super-resolution via
[26] A. Sharma, D.W. Jacobs, Bypassing synthesis: PLS for face recogni-
sparse representation, IEEE Trans. Image Process. 19 (11) (2010)
tion with pose, low-resolution and sketch, in: Proceedings of IEEE
28612873.
[12] H. He, W.C. Siu, Single image super-resolution using Gaussian Conference on Computer Vision and Pattern Recognition, 2011,
process regression, in: Proceedings of IEEE Conference on Computer pp. 593600.
Vision and Pattern Recognition, 2011, pp. 449456. [27] P. Chatterjee, P. Milanfar, A generalization of non-local means via
[13] K.I. Kim, Y. Kwon, Single-image super-resolution using sparse kernel regression, in: Proceedings of IS&T-SPIE Computational
regression and natural image prior, IEEE Trans. Pattern Anal. Mach. Imaging VI, 2008, p. 68140P.
Intell. 32 (6) (2010) 11271133. [28] I. Daubechies, M. Defrise, C. De Mol, An iterative thresholding
[14] H. Zhang, J. Yang, Y. Zhang, T.S. Huang, Non-local kernel regression algorithm for linear inverse problems with a sparsity constraint,
for image and video restoration, in: Proceedings of European Commun. Pure Appl. Math. 57 (11) (2004) 14131457.
Conference on Computer Vision, 2010, pp. 566579. [29] Z. Wang, A.C. Bovik, H.R. Sheikh, E.P. Simoncelli, Image quality
[15] G. Freedman, R. Fattal, Image and video upscaling from local self- assessment: from error visibility to structural similarity, IEEE Trans.
examples, ACM Trans. Graph. 28 (3) (2010) 110. Image Process. 13 (4) (2004) 600612.
[16] H. Takeda, S. Farsiu, P. Milanfar, Kernel regression for image proces- [30] W. Gao, B. Cao, S. Shan, X. Chen, D. Zhou, X. Zhang, D. Zhao, The CAS-
sing and reconstruction, IEEE Trans. Image Process. 16 (2) (2007) PEAL large-scale Chinese face database and baseline evaluations,
349366. IEEE Trans. Syst. Man Cybern. Part A 38 (1) (2008) 149161.
[17] A. Buades, B. Coll, J.M. Morel, A non-local algorithm for image [31] P.J. Phillips, H. Moon, S.A. Rizvi, P.J. Rauss, The FERET evaluation
denoising, in: Proceedings of IEEE Conference on Computer Vision methodology for face-recognition algorithms, IEEE Trans. Pattern
and Pattern Recognition, 2005, pp. 6065. Anal. Mach. Intell. 22 (10) (2000) 10901104.