The Gaussian Process Regression (GPR) framework offers a robust and intuitively probabilistic approach, presenting advantages over the less interpretable neural network models, particularly in the context of image super-resolution. Traditional applications of GPR in super-resolution have primarily focused on the posterior mean for generating predictions, neglecting the insightful predictive variance. This variance quantifies the confidence level of each prediction, a critical aspect for ensuring the reliability of the reconstructed images. In this study, we leverage this underutilized metric by incorporating a confidence regularization term into the iterative back projection process. This innovation aims to enhance the reconstruction fidelity by guiding the enhancement process in a more informed manner. Our extensive experimental analysis on established benchmark datasets confirms the superior performance and efficiency of our proposed approach, marking a significant advancement in super-resolution methodologies.
LiuDWangXFHanRD, et al.CTE-Net: contextual texture enhancement network for image super-resolution. IEEE Trans Multimed2024; 26(3): 8000–8013.
2.
ChenRYanZ. Learning dynamic generative attention for single image super-resolution. IEEE Trans Circ Syst Video Technol2022; 32(12): 8368–8382.
3.
ChungKLLeeYLChienWC. Effective gradient descent-based chroma subsampling method for Bayer CFA images in HEVC. IEEE Trans Circ Syst Video Technol2019; 29(11): 3281–3290.
4.
SongQXiongRLiuD, et al.Fast image super-resolution via local adaptive gradient field sharpening transform. IEEE Trans Image Process2018; 27(4): 1966–1980.
5.
DattaRMandalSUmerS, et al.Single-image reconstruction using novel super-resolution technique for large-scaled images. Soft Comput2022; 26(16): 8089–8103.
6.
ZhangYWangPBaoF, et al.A single-image super-resolution method based on progressive-iterative approximation. IEEE Trans Multimed2020; 22(6): 1407–1422.
7.
HuQWangXJiangJ, et al.Exploring the spectral prior for hyperspectral image super-resolution. IEEE Trans Image Process2024; 33(9): 5260–5272.
8.
MaCRaoYLuJ, et al.Structure-preserving image super-resolution. IEEE Trans Pattern Anal Mach Intell2022; 44(11): 7898–7911.
9.
LiangMYDuJPLiLH, et al.Video super-resolution reconstruction based on deep learning and spatio-temporal feature self-similarity. IEEE Trans Knowl Data Eng2022; 34(9): 4538–4553.
10.
BodtCMuldersDVerleysenM, et al.Fast multiscale neighbor embedding. IEEE Trans Neural Networks Learn Syst2022; 33(4): 1546–1560.
11.
WangYLiuWSunW, et al.A progressive feature enhancement deep network for large-scale remote sensing image super-resolution. IEEE Trans Geosci Rem Sens2023; 61(8): 1–13.
12.
WangHYChengSLLiYM, et al.Lightweight remote-sensing image super-resolution via attention-based multilevel feature fusion network. IEEE Trans Geosci Rem Sens2023; 61(11): 1–15.
13.
SuJNGanMChenGY, et al.High-similarity-pass attention for single image super-resolution. IEEE Trans Image Process2024; 33(1): 610–624.
14.
HuYLiJHuangYF, et al.Image super-resolution with self-similarity prior guided network and sample-discriminating learning. IEEE Trans Circ Syst Video Technol2022; 32(4): 1966–1985.
15.
KumarASinghHVKhareV. Tchebichef transform domain-based deep learning architecture for image super-resolution. IEEE Trans Neural Networks Learn Syst2024; 35(2): 2182–2193.
16.
HuangTYuanXDongW, et al.Deep Gaussian scale mixture prior for image reconstruction. IEEE Trans Pattern Anal Mach Intell2023; 45(9): 10778–10794.
17.
LuSPLiSMWangR, et al.Low-rank constrained super-resolution for mixed-resolution multiview video. IEEE Trans Image Process2021; 30(12): 1072–1085.
18.
MeiSJiangRLiX, et al.Spatial and spectral joint super-resolution using convolutional neural network. IEEE Trans Geosci Rem Sens2020; 58(7): 4590–4603.
19.
PanZMaWGuoJ, et al.Super-resolution of single remote sensing image based on residual dense backprojection networks. IEEE Trans Geosci Rem Sens2019; 57(10): 7918–7933.
20.
SharmaAShrivastavaBP. Different techniques of image SR using deep learning: a review. IEEE Sens J2023; 23(3): 1724–1733.
21.
ChenLPanJLiQ. Robust face image super-resolution via joint learning of subdivided contextual model. IEEE Trans Image Process2019; 28(12): 5897–5909.
22.
BarmanTDekaB. A deep learning-based joint image super-resolution and deblurring framework. IEEE Trans Artif Intell2024; 5(6): 3160–3173.
23.
DingQTLiangZYWangLG, et al.Not All Patches Are Equal: hierarchical dataset condensation for single image super-resolution. IEEE Signal Process Lett2023; 30(11): 1752–1756.
24.
ZhangMJXuJZhangJ, et al.SPH-Net: hyperspectral image super-resolution via smoothed particle hydrodynamics modeling. IEEE Trans Cybern2024; 54(7): 4150–4163.
25.
ZhengYJYangYSCheTT, et al.Image matting with deep Gaussian process. IEEE Trans Neural Networks Learn Syst2023; 34(11): 8879–8893.
26.
GengTLiuXYWangX, et al.Deep shearlet residual learning network for single image super-resolution. IEEE Trans Image Process2021; 30(4): 4129–4142.
27.
EsmaeilzehiAAhmadMOSwamyMNS. Ultralight-weight three-prior convolutional neural network for single image super resolution. IEEE Trans Artif Intell2023; 4(6): 1724–1738.
28.
LiJCFangFMLiJQ, et al.MDCN: multi-scale dense cross network for image super-resolution. IEEE Trans Circ Syst Video Technol2021; 31(7): 2547–2561.
29.
WangRXGongMMTaoDC. Receptive field size versus model depth for single image super-resolution. IEEE Trans Image Process2020; 29(9): 1669–1682.
30.
MaRDuX. A hardware-aware network for real-world single image super-resolutions. IEEE Trans Artif Intell2024; 5(7): 3482–3496.
31.
ZhangYTianYKongY, et al.Residual dense network for image restoration. IEEE Trans Pattern Anal Mach Intell2021; 43(7): 2480–2495.
32.
EsmaeilzehiAZaredarHHatzinakosD, et al.DPAN: a deep light-weight attention-based image super resolution network using multi-dimensional filter design technique. IEEE Signal Process Lett2023; 30(10): 1637–1641.
33.
KatzRDiabNBatenkovD. Decimated prony’s method for stable super-resolution. IEEE Signal Process Lett2023; 30(10): 1467–1471.
34.
AzaïsJMDe CastroYMourareauS. Testing Gaussian process with applications to super-resolution. Appl Comput Harmon Anal2020; 48(1): 445–481.
35.
LiuHOngYSShenX, et al.When Gaussian process meets big data: a review of scalable GPs. IEEE Trans Neural Networks Learn Syst2020; 31(11): 4405–4423.
36.
SchmidtAMorales-ÁlvarezPMolinaR. Probabilistic attention based on Gaussian processes for deep multiple instance learning. IEEE Trans Neural Networks Learn Syst2024; 35(8): 10909–10922.
37.
LuHMChenJSLiaoWC. Nonparametric regression via variance-adjusted gradient boosting Gaussian process regression. IEEE Trans Knowl Data Eng2021; 33(6): 2669–2679.
38.
ZhouFYaoRGLiuBZ, et al.Visual quality assessment for super-resolved images: database and method. IEEE Trans Image Process2019; 28(7): 3528–3541.
39.
WangHJNieYLYanB. Single-image super-resolution based on steering kernel and Gaussian process regression. Int J Pattern Recogn Artif Intell2021; 35(3): 2154006:1–2154006:16.
