图像视频压缩、点云数据压缩、图像处理

个人简介

施云惠，北京工业大学信息科学技术学院教授，博士生导师。2002年于哈尔滨工业大学仿真中心获得工学博士学位。2020年至2021年获得国家留学基金委资助在圣塔克拉拉大学（SCU）从事图像视频编码相关的访问研究。主要从事图像视频压缩、点云数据压缩和图像处理方向研究，在球面图像表示、球面图像小波变换、多维信号稀疏表示、鲁棒的稀疏表示等建模方法和屏幕视频编码及图像视频的深度学习压缩方面展开了深入研究。主持多项国家自然科学基金和北京市自然科学基金面上项目、863计划子课题、北京市科委项目等，参加多项国家自然科学基金重点项目，在TPAMI、TIP、TMM、TCSVT、CVPR、ACM MM 等重要国际期刊和会议上发表论文50多篇，获授权国家发明专利三十余项。

教育简历 

1997.03 – 2002.03 哈尔滨工业大学, 控制制导与仿真, 博士

1991.09 – 1994.03 哈尔滨工业大学, 计算数学, 硕士

1985.09 – 1989.07 黑龙江大学, 计算数学及其应用软件, 学士

工作履历

2024.06 – 至今  北京工业大学, 信息科学技术学院 教授

2016.09 – 2024.06 北京工业大学, 信息学部, 教授

2013.12 – 2016.09  北京工业大学, 城市交通学院, 教授

2010.12 – 2013.12  北京工业大学, 计算机学院, 教授

2003.09 – 2010.12  北京工业大学, 计算机学院, 副教授

2001.08 – 2003.09  哈尔滨工业大学, 数学系, 副教授

1994.03 – 2001.08  哈尔滨工业大学, 数学系, 讲师

学术兼职

1. 担任中国计算机学会多媒体技术专委会委员。

2. 担任TCSVT、TIP、TSP和JVCIR等多个国际期刊的审稿人。

课程教学

本科生教学：机器学习 数字信号处理 数据结构 编码原理 线性代数 计算方法

研究生教学：多媒体技术 图论

科研项目

1. 施云惠，规格化的球面图像表示及分析，国家自然科学基金，68.4万， 2020,01-2023,12

2. 施云惠，面向全景图像/视频的多维稀疏变换编码，国家自然科学基金，62万，2017,01 -2020,12

3. 施云惠，图像视频的压缩采样与稀疏重建的协同优化方法的研究，国家自然科学基金，73万，2014,01-2017,12

4. 施云惠，基于低秩表示的图像视频编码方法研究，国家自然科学基金，56万，2012,01-2015,12

5. 施云惠，基于压缩感知理论的图像/视频编解码技术研究，国家自然科学基金，31万2010,01-2012,12

6. 施云惠，基于视频多维描述空间的视频自适应技术研究(课题) ，国家高技术研究发展计划（863计划）41万，2006,06-2008,12

7. 施云惠，基于压缩感知理论的图像/视频编解码技术研究，北京市自然科学基金，11万2010,01-2012,12

8. 施云惠，珍稀档案保护和利用数字化关键技术研发与应用示范，北京市科委项目，165 万，2017,03- ⒛19,12

9. 施云惠，基于纹理合成的图像表示方法的研究,北京市教委项目,14万，2007,01-2009,12

10. 施云惠，交通视频大数据分析应用总体规划及关键技术研究，75万，2015,09-2015,12

荣誉和获奖

1．指导博士生获得国际会议IEEE VCIP 2016 Best Student Paper Award.

2．“多源高维数据协同表征及应用”获吴文俊人工智能科技进步奖一等奖。

代表性研究成果

1. 在多维信号表示上，提出一种多维合成/分析稀疏模型及高效的多维稀疏字典训练算法。该多维模型能够在保持多维信号原始结构下，通过一系列的字典刻画其各个维度的特性，高效表达多维信号。该稀疏表示模型(TPAMI, 2018；CVPR 2016。最新Google Scholar分别引用45次和47次)得到多位IEEE Fellow引用和正面评价。其中IEEE Fellow, C. L. Philip Chen教授评价“该工作是张量数据表示方法在机器学习领域应用的代表性工作”(IEEE TCYB 2018)。

2. 提出一种规格化的球面图像表示——基于球面测度的球面图像表示，克服传统的全景视频处理中采样的不均匀性和数据的不连续性带来的失真，为在保持球面图像结构研究全景图像视频的分析与压缩提供了崭新的思路。提出了基于球面测度的球面图像小波变换，为全景图像视频提供了一个多尺度分析工具。相应的研究成果发表在国际图像处理期刊IEEE TIP2021上。

3. 提出了超先验可变码率的图像压缩方法，建立了图像压缩的个体R-λ模型，该R-λ模型是第一个为可变码率深度学习图像压缩提供了连续码率控制的工作，实现了压缩中的精确和连续的码率控制。相应的研究成果在线发表在TCSVT 2022上。

4. 提出了基于预测-补偿的三维网格的对抗生成网，首次生成了具有真实感和多样性的三维人脸网格数据。提出了3D 网格的Wasserstein 质量评价方法，解决了传统的2D图像生成质量评价方法无法度量3D空间的三维网格生成质量的问题。相应的研究成果发表在TCSVT 2022。

主要论文论著

教材 《计算方法》, 张池平, 施云惠, 中国科学出版社, 2002

发表论文

Selected journal Papers

[1] Yunhui Shi, Kangfu Zhang, Jin Wang, Nam Ling ,Baocai Yin:Variable-Rate Image Compression Based on Side Information Compensation and R -λ Model Rate Control, IEEE Trans. Circuits Syst. Video Technol.  DOI 10.1109/TCSVT.2022.3232028

[2] Tingting Li, Yunhui Shi, Xiaoyan Sun, Jin Wang, Baocai Yin:PCGAN: Prediction-Compensation Generative Adversarial Network for Meshes. IEEE Trans. Circuits Syst. Video Technol. 32(7): 4667-4679 (2022)

[3] Jin Wang, Longhua Sun, Ruiqin Xiong, Yunhui Shi, Qing Zhu, Baocai Yin:Depth Map Super-Resolution Based on Dual Normal-Depth Regularization and Graph Laplacian Prior, IEEE Trans. Circuits Syst. Video Technol. 32(6): 3304-3318 (2022)

[4] Gang Wu,Yunhui Shi,Jin WangBaocai Yin: SMSIR: Spherical Measure Based Spherical Image Representation, IEEE Trans. on Image Processing . 30: 6377-6391 (2021)

[5] Zhenning Hou,Yunhui Shi,Jin Wang, Yingxuan Cui ,Baocai Yin: MS-Net: A lightweight Separable ConvNet for Multi-dimensional Image Processing,  Multimedia Tools and Applications. 80(17): 25673-25688 (2021)

[6] Jin Wang, Wei Xu, Jian-Feng Cai, Qing Zhu, Yunhui Shi, Baocai Yin: Multi-Direction Dictionary Learning Based Depth Map Super-Resolution With Autoregressive Modeling. IEEE Trans. Multim. 22(6): 1470-1484 (2020)

[7] Wenbin Yin, Yunhui Shi, Wangmeng Zuo, Xiaopeng Fan: A Co-Prediction-Based Compression Scheme for Correlated Images. IEEE Trans. Multim. 22(8): 1917-1928 (2020）

[8] Na Qi, Yunhui Shi, Xiaoyan Sun, Jingdong Wang, Baocai Yin, Junbin Gao:Multi-Dimensional Sparse Models. IEEE Trans. Pattern Anal. Mach. Intell. 40(1): 163-178 (2018)

[9] Xiaojie Liu, Wenpeng Ding, Yunhui Shi, Baocai Yin:Content adaptive interpolation filters based on HEVC framework. J. Visual Communication and Image Representation 56: 131-138 (2018)

[10] Wenbin Yin, Xiaopeng Fan, Yunhui Shi, Ruiqin Xiong, Debin Zhao:Compressive Sensing Based Soft Video Broadcast Using Spatial and Temporal Sparsity. MONET 21(6): 1002-1012 (2016)

[11] Weijia Zhu, Wenpeng Ding, Jizheng Xu, Yunhui Shi, Baocai Yin:Hash-Based Block Matching for Screen Content Coding. IEEE Trans. Multimedia 17(7): 935-944 (2015)

[12] Yunhui Shi, Bo Wen, Wenpeng Ding, Na Qi, Baocai Yin:Prediction-based realistic 3D model compression. Multimedia Tools Appl. 70(3): 2125-2137 (2014)

[13] Weijia Zhu, Wenpeng Ding, Jizheng Xu, Yunhui Shi, Baocai Yin: Screen Content Coding Based on HEVC Framework. IEEE Trans. Multimedia 16(5): 1316-1326 (2014)

[14] Zhen Zhang, Yunhui Shi, Wenpeng Ding, Baocai Yin:MR images reconstruction based on TVWL2-L1 model. J. Visual Communication and Image Representation 24(2): 187-195(2013)

[15] Jin Wang, Yunhui Shi, Dehui Kong, Wenpeng Ding, Chunjing Li, Baocai Yin:Sparse representation based down-sampling image compression. J. Computational Applied Mathematics 236(5): 675-683

[16] Zhen Zhang, Yunhui Shi, Dehui Kong, Wenpeng Ding, Baocai Yin:Image decoding optimization based on compressive sensing. J. Computational Applied Mathematics 236(5): 812-818 (2011)

[17] Wenpeng Ding, Ruiqin Xiong, Yunhui Shi, Dehui Kong, Baocai Yin:Fast mode dependent directional transform via butterfly-style transform and integer lifting steps. J. Visual Communication and Image Representation 22(8): 721-726 (2011)

Selected Conference Papers

[1] Jin Wang, Yunhui Shi, Yinsen Xing, Nam Ling, Baocai Yin:Deep Correlated Image Set Compression Based on Distributed Source Coding and Multi-Scale Fusion. DCC 2022: 192-201

[2] Yunhui Shi, Shaopei An, Jin Wang, Baocai Yin:Learned Bi-Directional Motion Prediction for Video Compression. MMAsia 2022: 15:1-15:6

[3] Huan Wang, Yunhui Shi, Jin Wang, Gang Wu, Nam Ling, Baocai Yin:Spherical Image Compression Using Spherical Wavelet Transform. MMAsia 2021: 47:1-47:5

[4] Yibo Wang, Jin Wang, Yunhui Shi, Longhua Sun, Baocai Yin:LGP-Net: Local Geometry Preserving Network for Point Cloud Completion. ICME 2022: 1-6

[5] Dongliang Shao, Yunhui Shi, Jin Wang, Nam Ling, Baocai Yin:A Model-Guided Unfolding Network for Single Image Reflection Removal. MMAsia 2021: 27:1-27:7

[6] Mengyun Liu, Na Qi, Yunhui Shi, Baocai Yin:An Attention Fusion Network For Event-Based Vehicle Object Detection. ICIP 2021: 3363-3367

[7] Michael G. Schimpf, Nam Ling, Yunhui Shi, Ying Liu: Sparse Coding of Intra Prediction Residuals for Screen Content Coding. ICCE 2021: 1-6

[8] Min Zhang, Yunhui Shi, Xiaoyan Sun, Nam Ling, Na Qi:Learning Redundant Sparsifying Transform based on Equi-Angular Frame. VCIP 2020: 439-442

[9] Jin Wang, Chen Wang, Qingming Huang, Yunhui Shi, Jian-Feng Cai, Qing Zhu, Baocai Yin:Image Inpainting Based on Multi-frequency Probabilistic Inference Model. ACM Multimedia 2020: 1-9

[10] Haoran Li, Wenpeng Ding, Yunhui Shi, Wenbin Yin: A Double Background Based Coding Scheme for Surveillance Videos. DCC 2018: 420

[11] Jianzhuo Su, Wenbin Yin, Yunhui Shi, Jin Wang, Wenpeng Ding: Collaborative Projection and Reconstruction Using Intrinsic Structure of Compressive Light Field Sensing. ICIP 2018: 2890-2894

[12] Na Qi, Yunhui Shi, Xiaoyan Sun, Baocai Yin:TenSR: Multi-dimensional Tensor Sparse Representation. CVPR 2016: 5916-5925

[13] Yuan Gao, Yunhui Shi, Shaofan Wang, Wenpeng Ding, Jin Wang, Baocai Yin:Realistic 3D Mesh Compression Based on Predicted Angle-Normal Images. DCC 2016: 595

[14] Xiaojie Liu, Wenpeng Ding, Yunhui Shi, Baocai Yin: Content Adaptive Interpolation Filters for HEVC Framework. DCC 2016: 618

[15] Shan Jia, Wenpeng Ding, Yunhui Shi, Baocai Yin: A fast sub-pixel motion estimation algorithm for HEVC. ISCAS 2016: 566-569

[16] Xiaolei Li, Cuiling Lan, Yunhui Shi, Wenpeng Ding, Xiaoyan Sun, Baocai Yin: Internal-video mode dependent directional transform. VCIP 2016: 1-4

[17] Xiuhuan Zang, Yunhui Shi, Jin Wang, Wenpeng Ding, Baocai Yin: Optimizing collaborative sparse dictionary for compressive light field photography. VCIP 2016:

[18] Na Qi, Yunhui Shi, Xiaoyan Sun, Wenpeng Ding, Baocai Yin: Single image super-resolution via 2D sparse representation. ICME 2015: 1-6

[19] Weijia Zhu, Wenpeng Ding, Jizheng Xu, Yunhui Shi, Baocai Yin: 2-D Dictionary Based Video Coding for Screen Contents. DCC 2014: 43-52

[20] Jin Wang, Jian-Feng Cai, Yunhui Shi, Baocai Yin: Incoherent dictionary learning for sparse representation based image denoising. ICIP 2014: 4582-4586

[21] Na Qi, Yunhui Shi, Xiaoyan Sun, Jingdong Wang, Wenpeng Ding: Two dimensional analysis sparse model. ICIP 2013: 310-314

[22] Na Qi, Yunhui Shi, Xiaoyan Sun, Jingdong Wang, Baocai Yin: Two dimensional synthesis sparse model. ICME 2013: 1-6

[23] Jin Wang, Yunhui Shi, Wenpeng Ding, Baocai Yin: Low rank recovery based image interpolation using local and nonlocal modeling. PCS 2013: 333-336

[24] Yunhui Shi, Bo Wen, Wenpeng Ding, Na Qi, Baocai Yin: Realistic mesh compression based on geometry image. PCS 2012: 133-136

[25] Weijia Zhu, Wenpeng Ding, Yunhui Shi, Yanfeng Sun, Baocai Yin: Adaptive intra modes reduction by clustering for H.264/AVC. ICIP 2011: 1665-1668