人脸超分辨率重建中投影空间的选择方法

张哲 1; 齐春 1; 张钊强 1; 陈晓璇 2

doi:10.7652/xjtuxb201808007

您当前的位置：

首页 >

文章列表页 >

人脸超分辨率重建中投影空间的选择方法

更新时间：2025-07-09

- 人脸超分辨率重建中投影空间的选择方法
- A Selection Method of Projection Space for Face Super-Resolution Reconstruction
- 西安交通大学学报 2018年52卷第8期页码：43-48
- 作者机构：
  
  西安交通大学电子与信息工程学院,西安,710049
- 作者简介：
- 基金信息：
  
  : 2017-11-16。作者简介: 张哲(1990—),男,博士生;齐春(通信作者),男,教授,博士生导师。基金项目: 国家自然科学基金资助项目(61572395,61675161)
- DOI：10.7652/xjtuxb201808007
  中图分类号： TP391.41
- 网络首发：2018-08-10，
  
  纸质出版：2018
- 稿件说明：
移动端阅览
张哲 1, 齐春 1, 张钊强 1, 等. 人脸超分辨率重建中投影空间的选择方法[J]. 西安交通大学学报, 2018,52(8):43-48.

A Selection Method of Projection Space for Face Super-Resolution Reconstruction[J]. 2018, 52(8): 43-48.
张哲 1, 齐春 1, 张钊强 1, 等. 人脸超分辨率重建中投影空间的选择方法[J]. 西安交通大学学报, 2018,52(8):43-48. DOI： 10.7652/xjtuxb201808007.

A Selection Method of Projection Space for Face Super-Resolution Reconstruction[J]. 2018, 52(8): 43-48. DOI： 10.7652/xjtuxb201808007.

摘要

针对人脸超分辨率重建中如何获得细节更为丰富的超分辨率重建结果问题

通过评估投影空间的一致性

给出了一种投影空间的选择方法。该方法首先根据图像样本空间与投影空间之间的映射关系计算高低分辨率图像样本所对应投影空间的投影值

然后随机选取若干对高低测试图像样本投影值作为重建目标

并通过邻域嵌入分别获取其对应的高低分重建权值

最后通过计算高低分重建权值间的余弦相似度

并进行直方图统计分析来评估投影空间的一致性并对投影空间进行选择。实验结果表明

该方法可以快速高效地对各投影空间进行评估与选择

其中优秀的投影空间能够将人脸超分辨率重建结果的峰值信噪比提升0.3 dB左右。

Abstract

A selection method of projection space based on assessing the consistency of projection spaces is proposed to achieve more detailed face super-resolution reconstruction results. Firstly

pairs of high and low resolutions are projected into a common space based on the mapping relationship between the original space and the projection space. Secondly

weights of reconstruction are calculated through a random selection of projected high and low resolution pairs to obtain cosine similarities. These cosine similarities are then used to evaluate and to choose projection spaces with histogramming approach. Experiment results show the efficiency of the proposed method. Moreover

with a cautiously selected projection space

the corresponding face super-resolution algorithm achieves about a 0.3 dB improvement on peak signal to noise ratio.

关键词

Keywords

references

BAKER S, KANADE T. Hallucinating faces [C]∥ Proceedings of IEEE International Conference on Automatic Face and Gesture Recognition. Piscataway, NJ, USA: IEEE, 2000: 83-88.

刘鑫, 张钊强, 姚佳文, 等. 低秩矩阵和结构化稀疏分解的视频背景差分方法 [J]. 西安交通大学学报, 2016, 50(6): 23-29.

LIU Xin, ZHANG Zhaoqiang, YAO Jiawen, et al. A background subtraction method based on decomposition of low rank and sparsity matrices [J]. Journal of Xi'an Jiaotong University, 2016, 50(6): 23-29.

王晓芳, 齐春. 一种运用显著性检测的行为识别方法 [J]. 西安交通大学学报, 2018, 52(2): 24-29, 44.

WANG Xiaofang, QI Chun. An action recognition method using saliency detection [J]. Journal of Xi'an Jiaotong University, 2018, 52(2): 24-29, 44.

陈晓璇, 齐春. 基于低秩矩阵恢复和联合学习的图像超分辨率重建 [J]. 计算机学报, 2014, 37(6): 1372-1379.

CHEN Xiaoxuan, QI Chun. Single-image super-resolution via low-rank matrix recovery and joint learning [J]. Chinese Journal of Computers, 2014, 37(6): 1372-1379.

CHANG H, YEUNG D Y, XIONG Y M. Super-resolution through neighbor embedding [C]∥ Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2004: 275-282.

LI B, CHANG H, SHAN S G, et al. Aligning coupled manifolds for face hallucination [J]. IEEE Signal Processing Letters, 2009, 16(11): 957-960.

LI B, CHANG H, SHAN S G. Locality preserving constraints for super-resolution with neighbor embedding [C]∥ Proceedings of IEEE International Conference on Image Processing. Piscataway, NJ, USA: IEEE, 2009: 1189-1192.

JIANG J J, HU R M. Locality-constraint iterative neighbor embedding for face hallucination [C]∥ Proceedings of IEEE International Conference on Multimedia and Expo. Piscataway, NJ, USA: IEEE, 2013: 1-6.

SU K, TIAN Q. Neighborhood issue in single-frame image super-resolution [C]∥ Proceedings of IEEE International Conference on Multimedia and Expo. Piscataway, NJ, USA: IEEE, 2005: 4-7.

GAO X B, ZHANG K B, TAO D C, et al. Joint learning for single image super resolution via coupled constraint [J]. IEEE Transactions on Image Processing, 2012, 21(2): 469-490.

HAO Y H, QI C. Face hallucination based on modified neighbor embedding and global smoothness constraint [J]. IEEE Signal Processing Letters, 2014, 21(10): 1187-1191.

SHI J G, QI C. Kernel-based face hallucination via dual regularization priors [J]. IEEE Signal Processing Letters, 2015, 22(8): 1189-1193.

PANG H J, GAO G W, JING X Y, et al. Kernel locality-constrained adaptive iterative neighbor embedding for face hallucination [C]∥ Proceedings of IEEE International Conference on Wireless Communications Signal Processing. Piscataway, NJ, USA: IEEE, 2016: 1-5.

CHEN X X, QI C. Nonlinear neighbor embedding for single image super-resolution via kernel mapping [J]. Signal Process, 2014, 94: 6-22.

CHEN L, HU R M, HAN Z, et al. Face super resolution based on parent patch prior for VLQ scenarios [J]. Multimedia Tools and Applications, 2017, 76(7): 10231-10254.

LI S H, LUO L K, PENG H. Low-rank representation for single image super-resolution using metric learning [C]∥ Proceedings of IEEE International Conference on Computer Science Education. Piscataway, NJ, USA: IEEE, 2017: 415-418.

HUANG H, HE H T, FAN X, et al. Super-resolution of human face image using canonical correlation analysis [J]. Pattern Recognition, 2010, 43(7): 2532-2543.

RAHIMAN V A, GEORGE S N. Single image super resolution using neighbor embedding and statistical prediction model [J]. Computers Electrical Engineering, 2017, 62: 281-292.

JIANG J J, MA X, CHEN C, et al. Single image super-resolution via locally regular-ized anchored neighbor-hood regression and nonlocal means [J]. IEEE Transactions on Multimedia, 2017, 19(1): 15-26.

GAO W, CAO B, SHAN S G, et al. The CAS-PEAL large-scale Chinese face database and baseline evaluations [J]. IEEE Transactions on Systems, Man, and Cybernetics: Part A Systems and Humans, 2008, 38(1): 149-161.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

全局重建和位置块残差补偿的人脸图像超分辨率算法

利用位置字典对的人脸图像超分辨率方法

采用上下文双自相关注意力的偏振图像超分辨率重建网络

自适应估计模糊参数的最大后验概率超分辨率复原算法

多尺度区域注意力InfoGAN车牌识别网络