I am now an Associate Professor (mater tutor) at the Department of Electronic Science and Technology, National University of Defense Technology (NUDT), China. Prior to being an academic scholar, I obtained my Ph.D degree at Imperial College London in 2020; finished my Msc.Eng and B.Eng in NUDT, China. My research interest lies in low-level image processing and intelligent signal processing, combining advanced non-convex optimization theory and statistical modeling using learning-aided approaches.
May 2024:Happy to announce that our paper Blind Super-Resolution Via Meta-Learning and Markov Chain Monte Carlo Simulation got accepted to TPAMI!
Mar 2024:Happy to announce that our paper A Dynamic Kernel Prior Model for Unsupervised Blind lmage Super-Resolution got accepted to CVPR!
Sep 2023:Happy to announce that our paper Meta-learning based Domain Prior with Application to Optical-lSAR lmage Translation got accepted to TCSVT!
Sep 2023:Happy to announce that our paper Metalearning-based alternating minimization algorithm for nonconvex optimization got accepted to TNNLS!
A Dynamic Kernel Prior Model for Unsupervised Blind Image Super-Resolution
Zhixiong Yang, Jingyuan Xia, Shengxi Li, Xinghua Huang, Shuanghui Zhang, Zhen Liu, Yaowen Fu, Yongxiang Liu
CVPR, 2024
Paper/
PDF/
Code
The proposed DKP is a plug-and-play kernel estimation tool, which can be combined with the off-the-shelf image restoration model, e.g., DIP and Diffusion model, to realize unsupervised blind image super-resolution.
Meta-learning based blind image super-resolution approach to different degradations
Zhixiong Yang and Jingyuan Xia and Shengxi Li and Wende Liu and Shuaifeng Zhi and Shuanghui Zhang and Li Liu and Yaowen Fu and Deniz Gündüz
Neural Networks, 2024
Paper/
PDF/
Code
DDSR is an unsupervised blind SISR method, which can handel different degradations, such as partial covered, noise, and dark.
Blind Super-Resolution Via Meta-Learning and Markov Chain Monte Carlo Simulation
Xia, Jingyuan and Yang, Zhixiong and Li, Shengxi and Zhang, Shuanghui and Fu, Yaowen and Gündüz, Deniz and Li, Xiang
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024
Paper/
PDF/
Code
This paper proposes a plug-and-play blind image super-resolution method based on the Meta-learning and Markov Chain Monte Carlo, which contributes to preventing bad local optimal solutions from the optimization perspective.
Meta-learning based Domain Prior with Application to Optical-ISAR Image Translation
Liao, Huaizhang and Xia, Jingyuan and Yang, Zhixiong and Pan, Fulin and Liu, Zhen and Liu, Yongxiang
IEEE Transactions on Circuits and Systems for Video Technology, 2023
Paper/
PDF/
Code
The proposed MLDP aims to realize the image translation from optical domain to ISAR domain. To generate realistic ISAR images, MLDP learns the scattering distribution features and the classifcation identifying features in physical and task perspectives. Meanwhile, MLDP applies the meta-learning strategy to improve its generalization ability to generate ISAR images under limited training samples.
Metalearning-Based Alternating Minimization Algorithm for Nonconvex Optimization
Xia, Jing-Yuan and Li, Shengxi and Huang, Jun-Jie and Yang, Zhixiong and Jaimoukha, Imad M. and Gündüz, Deniz
IEEE Transactions on Neural Networks and Learning Systems, 2023
Paper/
PDF/
Code
This paper proposes a meta-learning-based alternating minimization (MLAM) method for nonconvex problems of multiple variables, which aims to minimize a part of the global losses over iterations instead of carrying minimization on each subproblem, and it tends to learn an adaptive strategy to replace the handcrafted counterpart resulting in advance on superior performance.
Localized Incomplete Multiple Kernel k-Means With Matrix-Induced Regularization
Li, Miaomiao and Xia, Jingyuan and Xu, Huiying and Liao, Qing and Zhu, Xinzhong and Liu, Xinwang
IEEE Transactions on Cybernetics, 2023
Paper/
PDF
An improved Localized incomplete multiple kernel k -means (LI-MKKM), called LI-MKKM with matrix-induced regularization (LI-MKKM-MR), is proposed by incorporating a matrix-induced regularization term to handle the correlation among base kernels. We theoretically show that the local kernel alignment is a special case of its global one with normalizing each base kernel matrices.
A Learning-Aided Flexible Gradient Descent Approach to MISO Beamforming
Yang, Zhixiong and Xia, Jing-Yuan and Luo, Junshan and Zhang, Shuanghui and Gündüz, Deniz
IEEE Wireless Communications Letters, 2022
Paper/
PDF/
Code
This letter proposes a learning aided gradient descent (LAGD) algorithm to solve the weighted sum rate (WSR) maximization problem for multiple-input single-output (MISO) beamforming. The proposed LAGD algorithm directly optimizes the transmit precoder through implicit gradient descent based iterations, at each of which the optimization strategy is determined by a neural network, and thus, is dynamic and adaptive.
Miscellanea
meta reviewer:PRCV
reviewer:CVPR、NIPS、ICCV、ECCV、AAAI
Thank Dr. Jon Barron for sharing the source code of the website.