About

Bihan Wen

 

PhD Candidate

Electrical and Computer Engineering

University of Illinois at Urbana-Champaign

Coordinated Science Lab, 1308 W. Main Street, Urbana Illinois 61801

Email: bwen3 at illinois dot edu

News

  • April, 2018

  • [Publication] One paper got accepted to IJCAI 2018 (rate = 20%) [ Preprint ], [ codes ]. Congratulations to Ding, Xianming, and Atlas!

  • Mar, 2018

  • [Invited Talk] I am invited to give a guest lecture at Texas A&M University about computer vision applications using learning. [ News ]

    [Codes & Website] I released the basic Unitary Transform Learning code that was used in STROLLR paper (ICASSP 2017).

    [Publication] One paper got accepted to ICME 2018 as oral presentation (rate = 15%). Congratulations to Guan-Ming!

    [Invited Talk] I am invited to give a talk at Texas A&M University. [ News ]

    [Invited Talk] I am invited to present at Midwest Research Summit organized by Huawei [Best Presentation Award]

  • Feb, 2018

  • [Invited Talk] I am invited to give a talk on CSL Student Conference 2018. [ News ], [Best Talk Award judged and sponsored by NVIDIA]

    [Publicity & Service] I became the Reviewer for ICIP 2018.

    [Codes & Website] I released the Matlab codes of the FRIST algorithm (IVP 2017, ICIP 2016).

    [Publicity & Service] I became the Reviewer for ISCAIE 2018.

    [Invited Talk] I am invited to give a talk to Toutiao (今日头条).

  • Jan, 2018

  • [Publication] One paper got accepted to ICASSP 2018. [ PDF ] . Congratulations to Ulugbek, Dehong, Hassan, and Petros!

    [Publicity & Service] I joined the Reviewer Committee of Current Medical Imaging Reviews.

    [Codes & Website] I released the Matlab codes for the STROLLR denoising (ICASSP 2017) with state-of-the-art performance.

    [Publication] Research Summary I : A curated collection of reproducible image denoising methods .

  • Dec, 2017

  • [Publicity & Service] I joined the Program Committee of CVPR 2018.

    [Publication] One paper on VIDOSAT video denoising has been submitted to TIP.

  • Nov, 2017

  • [Codes & Website] Several codes are released in my Github page. Check out my Codes.

    [Publication] One paper on Image Forgery Detection has been accepted in ICSIP 2017. Congrats to Zhu Ye, T-T. Ng, X. Shen and B. Li!

  • Oct, 2017

  • [Publication] One paper has been accepted in ICCV 2017. Congratulations to Yanjun and Luke!

    [Publicity & Service] I become the reviewer for Journal of Electronic Imaging (JEI).

    Education

  • University of Illinois at Urbana-Champaign (UIUC), USA

  •      2012.8 - present, Master / PhD, Electrical and Computer Engineering (ECE)

         Advisor: Prof. Yoram Bresler

         Research Interests:

         Machine Learning, Sparse / Low-Rank Representation, Image / Video Processing,

         Computer Vision, Computational Imaging

  • Nanyang Technological University (NTU), Singapore

  •      2008.8 - 2012.6, Bachelor, Electrical and Electronic Engineering (EEE)

         Advisor: Prof. Yilong Lu

     

    Publications

  • Preprint

  •       B. Wen, S. Ravishankar, and Y. Bresler, “VIDOSAT - High-dimensional Sparsifying Transform Learning for Online Video Restoration,” IEEE Transcations on Image Processing (TIP), submitted. [ Arxiv:1710.00947 ]

  • 2018

  •       D. Liu, B. Wen, X. Liu, and T. Huang, “When Image Denoising Meets High-Level Vision Tasks: A Deep Learning Approach,” International Joint Conference on Artificial Intelligence (IJCAI), 2018, to appear. [ Arxiv:1706.04284 ], [ codes ]

          B. Wen, and G. Su, “TransIm: Transfer Image Local Statistics Across EOTFs for HDR Image Applications,” IEEE International Conference on Multimedia and Expo (ICME), 2018, to appear.

          B. Wen, U. Kamilov, D. Liu, H. Mansour, and P. Boufounos, “DeepCASD: An End-to-End Approach for Multispectral Image Super-Resolution,” IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018. [ PDF ]

  • 2017

  •       B. Wen, Y. Li, L. Pfister, and Y. Bresler, “Joint Adaptive Sparsity and Low-rankness on the Fly: An Online Tensor Reconstruction Method for Video Denoising,” IEEE International Conference on Computer Vision (ICCV), 2017. [ open access ], [ PDF ], [ codes ]

          B. Wen, S. Ravishankar, and Y. Bresler, “FRIST Flipping and Rotational Invariant Sparsifying Transform Learning and Applications,” Inverse Problems (IVP), 2017. [ PDF ], [ codes ]

          B. Wen, Y. Li, and Y. Bresler, “When Sparsity meets Low-Rankness: Transform Learning with Non-local Low-rank Constraint for Image Restoration,” IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2017. [ PDF ], [ codes ]

          Y. Zhu, T-T. Ng, B. Wen, X. Shen, and B. Li, “Copy-move Forgery Detection in the Presence of Similar but Genuine Objects,” IEEE International Conference on Signal and Image Processing (ICSIP), 2017. [ PDF ]

  • 2016

  •       B. Wen, S. Ravishankar, and Y. Bresler, “Structured Overcomplete Sparsifying Transform Learning with Convergence Guarantees and Applications,” International Journal of Computer Vision (IJCV), 2016. [ PDF ], [ codes ]

          D. Liu, Z. Wang B. Wen, J. Yang, W. Han, and T. Huang, “Robust Image Super-Resolution via Deep Networks with Sparse Prior,” IEEE Transcations on Image Processing (TIP), 2016. [ PDF ], [ codes ]

          S. Dev B. Wen, Y. Lee, and S. Winkler, “Ground-based Image Analysis: A Tutorial on Machine-Learning Techniques and Applications,” IEEE Geoscience and Remote Sensing Magazine (GRSM), 2016. [ PDF ]

          B. Wen, S. Ravishankar, and Y. Bresler, “Learning Flipping and Rotational Invariant Sparsifying Transform,” IEEE International Conference on Image Processing (ICIP), 2016. [ PDF ], [ codes ]

          B. Wen, Y. Zhu, R. Subramanian, T-T. Ng, X. Shen, and S. Winkler, “COVERAGE - A Novel Database for Copy-Move Forgery Detection,” IEEE International Conference on Image Processing (ICIP), 2016. [ PDF ], [ dataset ]

  • 2015

  •       S. Ravishankar, B. Wen, and Y. Bresler, “Online Sparsifying Transform Learning Part I: Algorithms,” IEEE Journal of Selected Topics in Signal Processing (JSTSP), 2015. [ PDF ]

          B. Wen, S. Ravishankar and Y. Brelser, “Video Denoising by Online 3D Sparsifying Transform Learning,” to appear in Proc. IEEE Int. Conf. Image Processing (ICIP), 2015. [ PDF ], , [ codes ]

  • 2014

  •       B. Wen, S. Ravishankar and Y. Brelser, “Learning Overcomplete Sparsifying Transforms with Block Cosparsity,” in Proc. IEEE Int. Conf. Image Processing (ICIP), 2014.  (10% Top Paper) [ PDF ], [ codes ]

          S. Ravishankar, B. Wen and Y. Brelser, “Online Sparsifying Transform Learning for Big Data Signal Processing,” in Proc. IEEE Global Conf. on Signal and Information Processing (Global SIP), 2014.  [ PDF ]

  • Undergrat Works

  •       B. Wen and Y. Lu, “A study of synthetic aperture radar imaging with compressed sensing,” in Proc. IEEE Asia-Pacific Conf. on Antennas and Propagation (APCAP), 2012.  [LINK]

          B. Wen and Y. Lu, “MATLAB tools for EnviSAT ASAR data visualization and image enhancement,” in Proc. SPIE Int. Symp. on Lidar and Radar Mapping Tech., 2011.  [LINK]

     

    Academic Services

  • Conference PC / Reviewer

  • IEEE International Conference on Image Processing (ICIP 2018)

    Conference on Computer Vision and Pattern Recognition (CVPR 2018)

    IEEE International Conference on Image Processing (ICIP 2017)

  • Journal Reviewer

  • IEEE Transactions on Image Processing ( TIP )

    IEEE Transactions on Signal Processing ( TSP )

    IEEE Transactions on Circuits and Systems for Video Technology ( TCSVT )

    IEEE Transactions on Information Forensics and Security ( TIFS )

    IEEE Transactions on Computational Imaging( TCI )

    Elsevier Neurocomputing Neurocomputing

    Electronics Letters ( EL )

    IET Radar, Sonar and Navigation ( RSN )

    Journal of Electronic Imaging ( JEI )

    Current Medical Imaging Reviews ( CMIR )

  • Conference Organization

  • Session Chair, Coordinated Science Laboratory Student Conference ( CSLSC 2017 )

    Session Assistant, Allerton Conference on Communication, Control, and Computing ( Allerton 2017 )

    Experiences

    Coordinated Science Lab

    Aug 2012 – present

    I am working with Prof. Yoram Bresler on machine learning for sparse data representation. We developed novel sparse model and transform learning algorithms for many data applications. We are working on extension to high-dimensional and large-scale problems, as well as combination with high-level vision problems with deep learning.

    Mitsubishi Elec. Research Lab. (MERL)

    May – Aug 2017

    I worked with Prof. Ulugbek Kamilov and Dr. Dehong Liu on multi-spectral data fusion problems. We proposed novel deep neural networks by unfolding the coupled dictionary learning.

    Dolby Laboratories

    May - Aug 2016

    I worked with Dr. Guan-Ming Su and Dolby Vision team on high dynamic range (HDR) video enhancement and reshaping problems. Our proposed methods have been patented and used in Dolby Vision codec.

    Advancec Digital Science Center

     

    May - Aug 2015

    I worked with Dr. Stefan and Dr. Rama Ratnam on several computer vision projects, including (i) machine learning applications in remote sensing, and (ii) image forgery detection using multi-model signal and data.

    ECE Department ECE 210

     

    2013 - 2014

    I was the head teaching assistant for ECE 210 Analog Signal Processing under Prof. Erhan Kudeki. I teached in the weekly lab and provided office hours. It is the major course for ECE undergraduate students. I was nominated for Harold L. Olesen Award by the students.

    Nanyang Technological University

    2008 - 2012

    I received my B.S. degree from Nanyang Technological University (NTU), Electrical and Electronic Engineering (EEE) in Singapore. During the 4 years undergraduate study, I worked with Prof. Yilong Lu for synthetic aperture radar (SAR) data analysis research.

    Sofewares

    Video Denoising via SALT

    The proposed algorithm denoises video online by reconstructing Sparse And Low-rank Tensor (SALT) sequentially.

    It Provides state-of-the-art denoising quality with cheap computation, small memory footprint, and low latency.

    Paper available [ here ].

    Download the codes [ here ].

    OCTOBOS learning and applications

    The proposed algorithm adaptively learns a structured overcomplete sparsifying transform with block cosparsity.

    It Provides simultaneous data clustering and reconstruction.

    Paper available [ here ].

    Download the codes [ here ].

    COVERAGE dataset for copy-move forgery detection

    COVERAGE contains copymove forged (CMFD) images and their originals with similar but genuine objects (SGOs).

    COVERAGE is designed to highlight and address tamper detection ambiguity of popular methods, caused by self-similarity within natural images.

    Paper available [ here ].

    Download the dataset [ here ].