Embed Size (px)
Citation preview
Tutorial on "Intelligent Image Enhancement and Restoration
-- from Prior-Driven Model to Advanced Deep Learning"
Proposal for IEEE ICME 2019
Instructors:
• Jiaying Liu, Associate Professor, Peking University
http://www.icst.pku.edu.cn/struct/
• Wenhan Yang, Postdoctoral Research Fellow, National University of Singapore
https://flyywh.github.io/
• Chen Change Loy, Associate Professor, Nanyang Technological University
http://www.ntu.edu.sg/home/ccloy/
Areas:
• Big multimedia data design and creation
• Deep learning for multimedia
• Artificial Intelligence for multimedia
• Brain-Inspired technologies for multimedia
• Prior-driven learning techniques
Keywords:
Prior model, deep learning, image restoration and enhancement, rain streak/drop removal,
low-light enhancement, super-resolution, large-scale datasets, statistic model
Abstract:
Intelligent image/video editing is a fundamental topic in image processing which has witnessed
rapid progress in the last two decades. Due to various degradations in the image and video capturing,
transmission and storage, image and video include many undesirable effects, such as low resolution,
low light condition, rain streak and rain drop occlusions. The recovery of these degradations is ill-
posed. With the wealth of statistic-based methods and learning-based methods, this problem can be
unified into the cross-domain transfer, which cover more tasks, such as image stylization.
In our tutorial, we will discuss recent progresses of image stylization, rain streak/drop removal,
image/video super-resolution, and low light image enhancement. This tutorial covers both
traditional statistics based and deep-learning based methods, and contains both biological-driven
model, i.e. Retinex model, and data-driven model. An image processing viewpoint that considers
the popular deep networks as a traditional Maximum-a-Posteriori (MAP) Estimation is provided.
The side priors, designed by researchers and learned by multi-task learnings, and automatically
learned priors, captures by adversarial learning are two kinds of important priors in this framework.
Three works under this framework, including single image super-resolution, low light image
enhancement, and single image raindrop removal are presented.
Single image super-resolution is a classical problem in computer vision. It aims at recovering a high-
resolution image from a single low-resolution image. This problem is an underdetermined inverse
problem, of which solution is not unique. In this tutorial, we will discuss how we can solve the
problem by deep convolutional networks in a data-driven manner. We will review different model
variants and important techniques such as adversarial learning for image super-resolution. We will
then discuss recent work on hallucinating faces of unconstrained poses and with very low resolution.
Finally, the tutorial will discuss challenges of implementing image super-resolution in real-world
scenarios.
Full Description (four pages)
1. Learning Objectives
The learning objectives of this tutorial includes four parts:
1) After the tutorial, the audiences can generally understand and follow the state-of-the-art
methods of related fields, their classifications, sources, and some potential future directions.
2) Let the audiences know the related applications of current state-of-the-art image restoration
methods, their capacities, potentialities and limitations.
3) Showing the audiences ways of building deep-learning image processing and restoration
methods or the applications in their domains based on existing well-known prior models or
bio-inspired models.
4) The audiences can make use of the related technologies to further facilitate their researches,
improving the state-of-the-art performance, boosting the robustness and benefiting practical
applications.
2. Outline
9:00-9:50
Summarization: Recent Progresses on Intelligent Image Restoration and Enhancement
- Super-Resolution
- Stylization
- Rain Streak/Drop Removal
- Low Light Enhancement
- A Prior Learning Framework: from Side Prior to Learned Prior
10:05-10:55
Tea Break & Off-line Communication
Prior Learning-Based Intelligent Image Restoration: Specific Applications
- Sub-Band Recovery-Guided Deep Image Super-Resolution
- Sparse Gradient Regularized Deep Retinex Model for Low Light Image Enhancement
- Attentive Generative Adversarial Network for Image Raindrop Removal
10:55-11:10
Tea Break & Off-line Communication
11:10-12:00
Single Image Super-Resolution by Deep Learning
- Single Image Super-Resolution by Convolutional Neural Networks
- Image Super-Resolution by Adversarial Learning
- Face Hallucination
3. Detailed Modules
3.1 Recent Progresses on Intelligent Image Restoration and Enhancement
In this part, we will briefly review the recent progress in the related fields of image restoration
and enhancement, including super-resolution, stylization, rain streak/drop removal, low light
enhancement.
3.2 A Prior Learning Framework: from Side Prior to Learned Prior
An image processing viewpoint that considers the popular deep networks as a traditional
Maximum-a-Posteriori (MAP) Estimation is provided. The priors can be considered as embedded
learned predictors having certain prior or biological meanings, or certain signal structures, where
traditional domain knowledge is incorporated as the constraints, such Retinex model, or sparse
coding. The deep networks become wiser by learning some potentially vital factors (side priors)
designed by the researchers. The recently proposed generative adverbial networks (GANs) provide
the capacity to automatically model the signal distributions and find the important elements (learned
priors) that might be useful to constrain the recovery of the degraded image into a clean one. With
this framework, we explore to design multi-task learning networks or regularized GAN networks to
address the low-level image enhancement problems.
3.3 Applications of Prior Learning-Based Intelligent Image Restoration
3.3.1 Sub-Band Recovery-Guided Deep Image Super-Resolution
In this work, we design a Deep Edge Guided REcurrent rEsidual (DEGREE) network that deeply
roots in traditional band recovery theory for single image super-resolution. DEGREE investigates
to recover the difference between a pair of LR and HR images by recurrent residual learning, and
further augments the SR process with edge-preserving capability. This work offers an understanding
on DEGREE from the view-point of sub-band frequency decomposition on image signal.
Reference:
[1] Wenhan Yang, Jiashi Feng, Jianchao Yang, Fang Zhao, Jiaying Liu, Zongming Guo, and Shuicheng Yan. "Deep
Edge Guided Recurrent Residual Learning for Image Super-Resolution", TIP, 2017.
[2] Ding Liu, Zhaowen Wang, Jianchao Yang, Wei Han, Thomas Huang, "Robust Single Image Super-Resolution
via Deep Networks with Sparse Prior," TIP, 2016.
[3] W. Lai, J. Huang, N. Ahuja and M. Yang, "Fast and Accurate Image Super-Resolution with Deep Laplacian
Pyramid Networks," TPAMI, 2018.
[4] Qi Mao, Shiqi Wang, Shanshe Wang, Xinfeng Zhang, Siwei Ma, "Enhanced Image Decoding via Edge-
Preserving Generative Adversarial Network", ICME, 2018.
3.3.2 Sparse Gradient Regularized Deep Retinex Model for Low Light Image Enhancement
In this work, we combine the advantages of Retinex model and deep learning method to address
the problem of low-light image enhancement, even in the presence of intensive noises, compression
artifacts, and their interleaved artifacts. We first construct a large-scale pioneering paired dataset
with real photography. Then, we build a deep network that works in the Retinex domain. The l0
gradient minimization is incorporated into the network to help extract the illumination map.
Combing the advantages of Retinex model and data-driven priors, the proposed network can not
only adjust the illumination, but also remove the noises and compression artifacts, and stretch the
contrast.
Reference:
[1] Wenhan Yang, Wenjing Wang, Jiaying Liu, Yuming Fang, "Sparse Gradient Regularized Deep Retinex Model for
Low Light Image Enhancement", Arxiv, 2018.
[2] Mading Li, Jiaying Liu, Wenhan Yang, Xiaoyan Sun, Zongming Guo, "Structure-Revealing Low-Light Image
Enhancement Via Robust Retinex Model", TIP, 2017.
[3] Xiaojie Guo, Yu Li, and Haibin Ling, "LIME: Low-light IMage Enhancement via Illumination Map Estimation"
TIP, 2017.
[4] Xueyang Fu, Delu Zeng, Yue Huang, Xiao-Ping Zhang, Xinghao Ding, "A Weighted Variational Model for
Simultaneous Reflectance and Illumination Estimation", CVPR, 2016.
3.3.3 Attentive Generative Adversarial Network for Image Raindrop Removal
In this work, we apply an attentive generative network using adversarial training to learn the prior
of raindrops, which further constrains and facilitates the raindrop removal. We inject visual attention
into both the generative and discriminative networks. During the training, our visual attention learns
about raindrop regions and their surroundings. Hence, by injecting this information, the generative
network will pay more attention to the raindrop regions and the surrounding structures, and the
discriminative network will be able to assess the local consistency of the restored regions.
Reference:
[1] Rui Qian, Robby Tan, Wenhan Yang, Jiajun Su, and Jiaying Liu. "Attentive Generative Adversarial Network for
Raindrop Removal from A Single Image", CVPR, 2018.
[2] Jiaying Liu, Wenhan Yang, Shuai Yang, and Zongming Guo. "D3R-Net: Dynamic Routing Residue Recurrent
Network for Video Rain Removal", TIP, 2018.
[3] Wenhan Yang, Robby T. Tan, Jiashi Feng, Jiaying Liu, Zongming Guo, and Shuicheng Yan. "Deep Joint Rain
Detection and Removal from a Single Image", CVPR, 2017.
[4] Xueyang Fu, Jiabin Huang, Xinghao Ding, Yinghao Liao, John Paisley, "Clearing the Skies: A deep network
architecture for single-image rain removal", TIP, 2017.
[5] He Zhang, Vishal M. Patel, "Density-aware Single Image De-raining using a Multi-stream Dense Network",
CVPR, 2018.
3.4 Single Image Super-Resolution by Deep Learning
3.4.1 Single Image Super-Resolution by Convolutional Neural Networks
Single image super-resolution (SISR), as a fundamental low-level vision problem, has attracted
increasing attention in the research community and AI companies. SISR aims at recovering a high-
resolution (HR) image from a single low-resolution (LR) one. Since the pioneer work of SRCNN
proposed by Dong et al. [1], deep convolution neural network (CNN) approaches have brought
prosperous development. Various network architecture designs and training strategies have
continuously improved the SR performance. In this session, we will discuss the merits and
drawbacks of various deep architectures that have been developed for SISR, e.g., SRCNN,
FSRCNN, VDSR, DRCN, LapSRN, DRRN, MemNet, SRDenseNet and EDSR.
References:
[1] C. Dong, C. C. Loy, K. He, and X. Tang, "Image Super-Resolution Using Deep Convolutional Networks," TPAMI,
2015.
[2] C. Dong‚ C. C. Loy, and X. Tang, "Accelerating the Super-Resolution Convolutional Neural Network," ECCV,
2016.
[3] J. Kim, J. K. Lee, K. M. Lee, "Accurate Image Super-Resolution Using Very Deep Convolutional Networks,"
CVPR, 2016.
[4] Y. Tai, J. Yang, X. Liu, "Image Super-Resolution via Deep Recursive Residual Network," CVPR, 2017.
[5] D. Liu, B. Wen, Y. Fan, C. C. Loy, T. S. Huang, "Non-Local Recurrent Network for Image Restoration," NIPS,
2018.
3.4.2 Image Super-Resolution by Adversarial Learning
Conventional SISR approaches typically optimize the models based on Peak Signal-to-Noise
Ratio (PSNR). PSNR-oriented approaches tend to output over-smoothed results without sufficient
high-frequency details. Several perceptual-driven methods have been proposed to improve the
visual
quality of SR results. For instance, perceptual loss is proposed to optimize super-resolution model
in a feature space instead of pixel space. Generative adversarial network is introduced to encourage
the network to favour solutions that look more like natural images. Semantic image prior is further
incorporated to improve recovered texture details. In this session, we will explain various techniques
that are based on adversarial learning for improving the image quality of SISR.
Reference:
[1] Christian Ledig, Lucas Theis, Ferenc Huszar, Jose Caballero, Andrew Cunningham, Alejandro Acosta, Andrew
Aitken, Alykhan Tejani, Johannes Totz, Zehan Wang, Wenzhe Shi, "Photo-Realistic Single Image Super-Resolution
Using a Generative Adversarial Network," CVPR, 2017.
[2] M. S. M. Sajjadi, B. Schölkopf, M. Hirsch, "Enhancenet: Single Image Super-Resolution Through Automated
Texture Synthesis," ICCV, 2017.
[3] X. Wang, K. Yu, C. Dong, and C. C. Loy, "Recovering Realistic Texture in Image Super-Resolution By Spatial
Feature Modulation," CVPR, 2018.
[4] Xintao Wang, Ke Yu, Chao Dong and Chen Change Loy, "ESRGAN: Enhanced Super-Resolution Generative
Adversarial Networks," PIRM Workshop, in conjunction with ECCV, 2018.
3.4.3 Face Hallucination
Increasing attention is devoted to detection of small faces with an image resolution as low as 10
pixels of height. Meanwhile, facial analysis techniques, such as face alignment and verification,
have seen rapid progress. However, the performance of most existing techniques would degrade
when given a low-resolution facial image, because the input naturally carries less information, and
images corrupted with down-sampling and blur would interfere the facial analysis procedure. Face
hallucination, a task that super-resolves facial images, provides a viable means for improving low-
res face processing and analysis, e.g. person identification in surveillance videos and facial image
enhancement. In this section we will cover recent techniques such as Deep Cascaded Bi-Network
for Face Hallucination, FSRNet, and Super-FAN. We will pay particular focus on how face prior
can be used to enhance the quality of face hallucination.
References:
[1] S. Zhu, S. Liu, C. C. Loy, and X. Tang, "Deep Cascaded Bi-Network for Face Hallucination," ECCV, 2016.
[2] A. Bulat, G. Tzimiropoulos, "Super-FAN: Integrated Facial Landmark Localization and Super-Resolution of
Real-World Low-Resolution Faces in Arbitrary Poses with GANS," CVPR, 2018.
[3] Yu Chen, Ying Tai, Xiaoming Liu, Chunhua Shen, Jian Yang, "FSRNet: End-to-End Learning Face Super-
Resolution with Facial Priors," CVPR, 2018.
4. Target Audiences
This tutorial reflects these recent developments while providing a comprehensive introduction to
the fields of image processing and deep learning. It is aimed at both undergraduates and PhD
students, as well as researchers and practitioners who are interested in deep learning, image
enhancement and restoration.
5. Prerequisite Knowledge Required
• Familiar with the basic concept of image processing.
• Knowing the basic idea of deep learning, and being familiar with typical network structures.
• Having the base of statistics (Optional).
• Understanding the basic convex optimization techniques (Optional).
6. Importance and Relevance of Tutorial
This tutorial tries to fill in the blank of the following issues:
1) A brief but comprehensive summarization of recent progresses in image enhancement and
restoration, including super-resolution, stylization, rain streak/drop removal and low light
enhancement.
2) Bridging the gap between prior and bio-inspired models and deep learning and illustrating ways
of borrowing from prior and bio-inspired models to construct deep learning pipelines.
3) Providing a systematic view of the relationship between traditional domain knowledge, i.e.
prior and bio-inspired models and deep learning.
4) An important complementary summarization to existing deep-learning or prior and bio-
inspired methods.
5) Providing more research tools to both low-level vision researchers and high-level computer
vision researchers, further improving the state-of-the-art performance, boosting the robustness
and benefiting practical applications.
Our tutorial has very close a relationship to many multimedia research fields. The idea of borrowing
prior and bio-inspired models to facilitate deep learning methods will benefit developing new
methods and applications in related multimedia research topics. Our tutorial will provide new ideas,
new perspectives, new tools and new methods to the community. It will be a great honor for us, if
we have the opportunity to convey these insights to others at the ICME-2019 tutorial.
Jiaying Liu
Ph. D, Associate Professor
Institute of Computer Science & Technology Email: [email protected]
Peking University http://www.icst.pku.edu.cn/struct/
No. 5 Yiheyuan Road, Haidian District Phone: 8610-82529714
Beijing, China, 100871 Fax: 8610-62754532
RESEARCH INTERESTS
Multimedia and signal processing with emphasis on Digital Image/Video Processing and
Compression, Computer Vision Techniques, including
• Image/Video Restoration and Enhancement
• Image Stylization and Assessment
• Action Understanding and Analytics
EDUCATION
• Peking University Beijing, China
Ph. D in Computer Science 07/2010
Thesis: Rate Distortion Optimization Based Scalable Video Coding
• Northwestern Polytechnic University Xi’an, Shaanxi, China
B. E. in Computer Science / Honor Class 06/2005
WORK EXPERIENCES
• Peking University Beijing, China
Associate Professor, Institute of Computer Science and Technology 08/2012 – Present
Assistant Professor, Institute of Computer Science and Technology 07/2010 – 07/2012
• Microsoft Research Asia Beijing, China
Visiting Researcher, Internet Multimedia Group 03/2015 – 10/2015
MSRA Star Track for Young Faculties Program
• University of Southern California, (Aug. 2007 – Aug. 2008) Los Angeles, CA, US
Visiting Scholar, Collaborate with Prof. C.-C. Jay Kuo 08/2007 – 08/2008
RELATED PUBLICATIONS
[1] Jiaying Liu, Wenhan Yang, Shuai Yang, and Zongming Guo. "D3R-Net: Dynamic Routing Residue
Recurrent Network for Video Rain Removal", Accepted by IEEE Trans. on Image Processing (TIP), Sep. 2018.
[2] Jiaying Liu, Wenhan Yang, Shuai Yang, Zongming Guo. "Erase or Fill? Deep Joint Recurrent Rain
Removal and Reconstruction in Videos", Proc. of IEEE Conference on Computer Vision and Pattern
Recognition (CVPR), Salt Lake City, Utah, U.S., June 2018.
[3] Rui Qian, Robby Tan, Wenhan Yang, Jiajun Su, and Jiaying Liu. "Attentive Generative Adversarial
Network for Raindrop Removal from A Single Image", Proc. of IEEE Conference on Computer Vision and
Pattern Recognition (CVPR), Salt Lake City, Utah, U.S., June 2018.
[4] Wenhan Yang, Robby T. Tan, Jiashi Feng, Jiaying Liu, Zongming Guo, and Shuicheng Yan. "Deep Joint
Rain Detection and Removal from a Single Image", Proc. of IEEE Conference on Computer Vision and Pattern
Recognition (CVPR), Honolulu, Hawaii, Jul. 2017.
[5] Wenhan Yang, Jiashi Feng, Jianchao Yang, Fang Zhao, Jiaying Liu, Zongming Guo, and Shuicheng Yan.
"Deep Edge Guided Recurrent Residual Learning for Image Super-Resolution", IEEE Trans. on Image
Processing (TIP), Vol.26, No.12, pp.5895-5907, Dec. 2017.
PREVIOUS TUTORIALS / TALKS
1. ChinaMM-2018
Jiaying Liu, Intelligent Image Enhancement Computing, http://tcmt.ccf.org.cn/chinamm/2018/jiangxiban.html
2. IEEE VCIP-2018
Jiaying Liu, Wenhan Yang, Intelligent Image/Video Editing, http://vcip2018.org/tutorials.php#T
3. PRCV-2018
Jiaying Liu, Jinshan Pan, Zhiwei Xiong, Intelligent Image Enhancement Computing
GRANTS
[1] PI: National Natural Science Foundation of China (NSFC), No. 61772043. "Research on Joint Video
Reconstruction and Analytics Under Severe Conditions", 2018.01—2021.12.
[2] PI: CCF-Tencent Rhino Bird Open Funding, CCF-Tencent RAGR20170108. "Deep Learning Based Video
Analytics and Understanding", 2017.10—2018.12.
[3] PI: MSRA Collaborative Research (MSRA-CR), "Remote Health and Medical Aide via Skeleton-Based
Action Forecast and Detection", Project ID FY17-RES-THEME-013, 2017.01—2017.12.
[4] PI: National Natural Science Foundation of China (NSFC), No.61472011. "Image Representation,
Sparse Coding and Reconstruction Based on Sparse Representation", 2015.01—2018.12.
[5] PI: Beijing Natural Science Foundation (BJNSF), No.4142021. "Research on Image Modeling and
Reconstruction Based on Sparse Representation", 2014.01—2016.12.
[6] PI of Sub-Project: National High Technology Research and Development Program of China (863
Program), 2014AA015205, "Key Technology of Information Consumption Service and Application
Demonstration Based on Media Bid Data", 2014.01—2016.12.
[7] PI: National Natural Science Foundation of China (NSFC), No.61101078, "Research on Omni-Scalable
Video Coding and Super-Resolution Reconstruction", 2012.01—2014.12.
HONORS AND AWARDS
• Excellent Teaching Award, Peking University, 2018
• WangXuan Young Faculty Award, Peking University, 2016
• Excellent Teaching Award, Peking University, 2015
• Higher Institution Excellent Science and Technology Achievement Award, Ministry of Education of China, 2012
• Best Doctoral Dissertation, Peking University, 2012
• Excellent Teaching Award for Young Faculties, Peking University, 2011
PROFESSIONAL ACTIVITIES
• IEEE/CCF, Senior Member
• IEEE CASS-MSA/EOT, Technical Committee Member
• APSIPA, IVM & FPC, Technical Committee Member
• CCF TCMT, Technical Committee Member & Secretary
• CSIG BVC, Technical Committee Member & Secretary
• APSIPA, Distinguished Lecturer, 2016-2017
• Publicity Chair, IEEE International Conference on Image Processing (ICIP), to be held in Taipei,
Taiwan, 2019
• Grand Challenge Chair, IEEE International Conference on Multimedia & Expo (ICME), to be held
in Shanghai, China, 2019
• Workshop Chair, IEEE International Conference on Multimedia Information Processing and
Retrieval (MIPR), to be held in San Jose, CA, USA, 2019
• Publicity Chair & Area Chair, IEEE International Conference on Visual Communications and
Image Processing ( VCIP), to be held in Taichung, Taiwan, 2018
• Area Chair & Special Session Co-organizer, IEEE International Conference on Multimedia & Expo
(ICME), to be held in San Diego, CA, USA, 2018
• Finance Chair, IEEE International Conference on Multimedia Big Data (BigMM), to be held in
Xi’an, Shaanxi, China, 2018
• Registration Chair, IEEE International Conference on Visual Communications and Image
Processing ( VCIP), St. Petersburg, FL, USA, 2017
• Workshop Co-organizer, Workshop on Large Scale 3D Human Activity Analysis Challenge in Depth
Videos, IEEE International Conference on Multimedia & Expo (ICME), Hong Kong, 2017
• Special Session Co-organizer, Special Session on Advanced Multimedia Processing and
Compression, IEEE International Conference on Visual Communications and Image Processing
( VCIP), Chengdu, Sichuan, China, 2016
• Special Session Co-organizer, Special Session on Advanced Video Processing, Assessment and
Analysis, APSIPA Annual Summit and Conference (ASC), Jeju, Korea, 2016
• Special Session Co-organizer, Special Session on Image Sparse Representation and Its
Applications, IEEE International Conference on Visual Communications and Image Processing
( VCIP), Singapore, 2015
• Special Session Co-organizer, Special Session on Image Restoration via Low-Rank Approach and
Transform Domain, APSIPA Annual Summit and Conference (ASC), Hong Kong, 2015
• Publication Chair & Special Session Co-organizer, APSIPA Annual Summit and Conference (ASC),
Hollywood, CA, USA, 2012
SELECTED PUBLICATIONS (* corresponding author)
[1] Jiaying Liu, Wenhan Yang, Shuai Yang, and Zongming Guo. "D3R-Net: Dynamic Routing Residue Recurrent
Network for Video Rain Removal", Accepted by IEEE Trans. on Image Processing (TIP), Sep. 2018.
[2] Jiaying Liu, Shuai Yang, Yuming Fang, Zongming Guo. "Structure-Guided Image Inpainting Using
Homography Transformation", Accepted by IEEE Transactions on Multimedia (TMM), April 2018.
[3] Jiaying Liu, Yanghao Li, Sijie Song, Junliang Xing, Cuiling Lan, and Wenjun Zeng. "Multi-Modality Multi-
Task Recurrent Neural Network for Online Action Detection", Accepted by IEEE Trans. on Circuit System
for Video Technology (TCSVT), Jan. 2018.
[4] Jiaying Liu, Wenhan Yang, Xiaoyan Sun, and Wenjun Zeng. "Photo Stylistic Brush: Robust Style Transfer
via Superpixel-Based Bipartite Graph", IEEE Trans. on Multimedia (TMM), Vol.20, No.7, pp.1724-1737,
July 2018.
[5] Jiaying Liu, Wenhan Yang, Xinfeng Zhang, and Zongming Guo. "Retrieval Compensated Group Structured
Sparsity for Image Super-Resolution", IEEE Trans. on Multimedia (TMM), Vol.19, No.2, pp.302-316, Feb.
2017.
[6] Jiaying Liu, Yongjin Cho, Zongming Guo and C.-C. Jay Kuo. “Bit Allocation for Spatial Scalability Coding
of H.264/SVC with Dependent Rate-Distortion Analysis”, IEEE Transactions on Circuit System for Video
Technology (TCSVT), Vol.20, No.7, pp.967-981, Jul. 2010.
[7] Jiaying Liu, Wenhan Yang, Shuai Yang, Zongming Guo. "Erase or Fill? Deep Joint Recurrent Rain Removal
and Reconstruction in Videos", Proc. of IEEE Conference on Computer Vision and Pattern
Recognition (CVPR), Salt Lake City, Utah, U.S., June 2018.
[8] Jie Ren, Jiaying Liu*, and Zongming Guo. "Context-Aware Sparse Decomposition for Image Denoising and
Super-Resolution", IEEE Transactions on Image Processing (TIP), Vol.22, No.4, pp.1456-1469, Apr. 2013.
[9] Wenhan Yang, Jiashi Feng, Jianchao Yang, Fang Zhao, Jiaying Liu*, Zongming Guo, and Shuicheng Yan.
"Deep Edge Guided Recurrent Residual Learning for Image Super-Resolution", IEEE Trans. on Image
Processing (TIP), Vol.26, No.12, pp.5895-5907, Dec. 2017.
[10] Yongqin Zhang, Jiaying Liu*, Wenhan Yang and Zongming Guo. "Image Super-Resolution Based on
Structure-Modulated Sparse Representation", IEEE Transactions on Image Processing (TIP), Vol.24, No.9,
pp.2797-2810, May 2015.
[11] Shuai Yang, Jiaying Liu*, Yuming Fang, and Zongming Guo. "Joint-Feature Guided Depth Map Super-
Resolution with Face Priors", Accepted by IEEE Transactions on Cybernetics (TCYB), Dec. 2016.
[12] Wenhan Yang, Jiaying Liu*, Mading Li, and Zongming Guo. "Isophote-Constrained Autoregressive Model
with Adaptive Window Extension for Image Interpolation", Accepted by IEEE Trans. on Circuit System for
Video Technology (TCSVT), Dec. 2016.
[13] Mading Li, Jiaying Liu*, Jie Ren, Zongming Guo. "Adaptive General Scale Interpolation Based on Weighted
Autoregressive Models". IEEE Transactions on Circuit System for Video Technology (TCSVT). Vol.25, No.2,
pp.200-211, Feb. 2015.
[14] Yuming Fang, Jiebin Yan, Jiaying Liu*, Shiqi Wang, Qiaohong Li, and Zongming Guo, "Objective Quality
Assessment of Screen Content Images by Uncertainty Weighting", IEEE Transactions on Image Processing
(TIP), Vol.26, No.4, Apr. 2017.
[15] Xinfeng Zhang, Ruiqin Xiong, Weisi Lin, Siwei Ma, Jiaying Liu and Wen Gao. "Video Compression Artifact
Reduction via Spatio-Temporal Multi-Hypothesis Prediction", IEEE Transactions on Image Processing (TIP),
Vol.24, No.12, pp.6048-6016, Dec. 2015.
[16] Jun Sun, Yizhou Duan, Jiangtao Li, Jiaying Liu*, and Zongming Guo. "Rate-Distortion Analysis of Dead-
Zone Plus Uniform Threshold Scalar Quantization and Its Application—Part I: Fundamental Theory", IEEE
Transaction on Image Processing (TIP), Vol.22, No.1, pp.202-214, Jan. 2013.
[17] Jun Sun, Yizhou Duan, Jiangtao Li, Jiaying Liu*, and Zongming Guo. "Rate-Distortion Analysis of Dead-
Zone Plus Uniform Threshold Scalar Quantization and Its Application—Part II: Two-Pass VBR Coding for
H.264/AVC", IEEE Transactions on Image Processing (TIP), Vol.22, No.1, pp.215-228, Jan. 2013.
[18] Shuai Yang, Jiaying Liu*, Zhouhui Lian, and Zongming Guo. "Awesome Typography: Statistics-Based Text
Effects Transfer", Accepted by IEEE Conference on Computer Vision and Pattern Recognition (CVPR),
Honolulu, Hawaii, Jul. 2017.
[19] Wenhan Yang, Robby T. Tan, Jiashi Feng, Jiaying Liu, Zongming Guo, and Shuicheng Yan. "Joint Rain
Detection and Removal from a Single Image", Accepted by IEEE Conference on Computer Vision and
Pattern Recognition (CVPR), Honolulu, Hawaii, Jul. 2017.
[20] Sijie Song, Cuiling Lan, Junliang Xing, Wenjun Zeng and Jiaying Liu*. "An End-to-End Spatio-Temporal
Attention Model for Human Action Recognition from Skeleton Data". Proc. of AAAI Conference on Artificial
Intelligence (AAAI), San Francisco, CA, Feb. 2017 (Full / Oral).
Wenhan YANG
Ph. D, Postdoctoral Research Fellow
CONTACT
Address Vision & Machine Learning Lab, Block E4 #08-24, 4 Engineering Drive 3,
National University of Singapore, Singapore 117583.
Mobile +86-15120052873
Wechat marsywh
Email [email protected]
Web https://flyywh.github.io/
RESEARCH INTEREST
Image and video super-resolution
Real-time video processing
Bad weather restoration
Joint task of low-level processing and high-level computer vision
EDUCATION
2012.9 – 2018.7 Ph.D. student in Computer Science.
Institute of Computer Science & Technology,
Peking University, China. Supervisor: Prof. Zongming Guo and Jiaying Liu.
2008.9 – 2012.7 B.S. major in Computer Science.
School of Electronics Engineering and Computer Science,
Peking University, China.
2009.9 – 2012.7 B.S. minor in Economy.
National School of Development, Peking University, China.
WORKING EXPERIENCE
2018.09 – present Postdoctoral research fellow in Learning and Vision (LV) Group,
Department of Electrical & Computer Engineering, National University of
Singapore, Singapore.
Director: Prof. Jiashi Feng.
Worked on semantic information-guided image processing, few-shot
image processing, Joint task of low-level processing and high-level
computer vision.
2018.07 – 2018.09 Researcher in STRUCT Group. Institute of Computer Science &
Technology, Peking University, China.
Collaborator: Prof. Jiaying Liu.
Worked on solving the inverse problem of the image restoration and
bad weather restoration.
2015.9 – 2016.9 Visiting scholar in Learning and Vision (LV) Group, Department of
Electrical & Computer Engineering, National University of Singapore,
Singapore.
Collaborator: Prof. Jiashi Feng and Prof. Shuicheng Yan.
Worked on deep learning-based image and video super-resolution,
rain removal.
HONORS AND PROFESSIONAL ACTIVITIES
Distinguished doctorate dissertation of Peking University, 2018.
Outstanding graduates, Peking University, 2018.
Candidate of Outstanding Research Award, Peking University, 2018.
First Runner-up of CVPR-2018 UG2 Challenge, 2018.
Outstanding Research Award, School of EECS, Peking University, 2017.
Excellent talents Award, Cooperative Medianet Innovation Center, 2017.
Founder Scholarship, Founder Group, 2017.
Merit Student, School of EECS, Peking University, 2017.
Excellent Paper Award, ICST, Peking University, 2017.
Excellent Student Award, ICST, Peking University, 2017.
Reviewer for
- IEEE Transactions on Image Processing
- IEEE Transactions on Circuits and Systems for Video Technology
- IEEE Transactions on Multimedia
- IEEE Access
- Neurocomputing
RELATED PUBLICATIONS
RELATED PUBLICATIONS
[1] Wenhan Yang, Robby T. Tan, Jiashi Feng, Jiaying Liu, Zongming Guo, and Shuicheng Yan.
"Deep Joint Rain Detection and Removal from a Single Image", Proc. of IEEE Conference on
Computer Vision and Pattern Recognition (CVPR), Honolulu, Hawaii, Jul. 2017.
[2] Wenhan Yang, Jiashi Feng, Jianchao Yang, Fang Zhao, Jiaying Liu, Zongming Guo, and
Shuicheng Yan. "Deep Edge Guided Recurrent Residual Learning for Image Super-Resolution", IEEE
Trans. on Image Processing (TIP), Vol.26, No.12, pp.5895-5907, Dec. 2017.
[3] Jiaying Liu, Wenhan Yang, Shuai Yang, and Zongming Guo. "D3R-Net: Dynamic Routing
Residue Recurrent Network for Video Rain Removal", Accepted by IEEE Trans. on Image Processing
(TIP), Sep. 2018.
[4] Jiaying Liu, Wenhan Yang, Shuai Yang, Zongming Guo. "Erase or Fill? Deep Joint Recurrent
Rain Removal and Reconstruction in Videos", Proc. of IEEE Conference on Computer Vision and
Pattern Recognition (CVPR), Salt Lake City, Utah, U.S., June 2018.
[5] Rui Qian, Robby Tan, Wenhan Yang, Jiajun Su, and Jiaying Liu. "Attentive Generative
Adversarial Network for Raindrop Removal from A Single Image", Proc. of IEEE Conference on
Computer Vision and Pattern Recognition (CVPR), Salt Lake City, Utah, U.S., June 2018.
PREVIOUS TUTORIALS / TALKS
IEEE VCIP-2018 Jiaying Liu, Wenhan Yang, Intelligent Image/Video Editing,
http://vcip2018.org/tutorials.php#T
SELECTED PUBLICATIONS
Journal
[1] Wenhan Yang, Jiashi Feng, Jianchao Yang, Fang Zhao, Jiaying Liu, Zongming Guo, and
Shuicheng Yan. "Deep Edge Guided Recurrent Residual Learning for Image Super-Resolution",
IEEE Trans. on Image Processing (TIP), Vol. 26, No. 12, pp. 5895-5907, Dec. 2017.
[2] Wenhan Yang, Sifeng Xia, Jiaying Liu, and Zongming Guo. "Reference Guided Deep Super-
Resolution via Manifold Localized External Compensation", Accepted by IEEE Trans. on
Circuit System for Video Technology (TCSVT), May 2018.
[3] Wenhan Yang, Jiaying Liu, Mading Li, and Zongming Guo. "Isophote-Constrained
Autoregressive Model with Adaptive Window Extension for Image Interpolation", IEEE Trans.
on Circuit System for Video Technology (TCSVT), Vol.28, No.5, pp.1071-1086, May 2018..
[4] Wenhan Yang, Jiashi Feng, Guosen Xie, Jiaying Liu, Zongming Guo, and Shuicheng Yan.
"Video Super-Resolution Based on Spatial-Temporal Recurrent Residual Networks", Computer
Vision and Image Understanding (CVIU), Vol.168, pp.79-92, March. 2018.
[5] Wenhan Yang, Jiaying Liu, Sifeng Xia, Zongming Guo. "Data-Driven External Compensation
Guided Deep Networks for Image Super-Resolution", Ruan Jian Xue Bao/Journal of Software,
Vol.29, No.4, pp.900-913, April 2018.
[6] Jiaying Liu, Wenhan Yang, Shuai Yang, and Zongming Guo. "D3R-Net: Dynamic Routing
Residue Recurrent Network for Video Rain Removal", Accepted by IEEE Trans. on Image
Processing (TIP), Sep. 2018.
[7] Jiaying Liu, Wenhan Yang, Xiaoyan Sun, and Wenjun Zeng. "Photo Stylistic Brush: Robust
Style Transfer via Superpixel-Based Bipartite Graph", Accepted by IEEE Trans. on Multimedia
(TMM), 2017.
[8] Jiaying Liu, Wenhan Yang, Xinfeng Zhang, and Zongming Guo. "Retrieval Compensated
Group Structured Sparsity for Image Super-Resolution", IEEE Trans. on Multimedia (TMM),
Vol.19, No.2, pp.302-316, Feb. 2017.
[9] Yongqin Zhang, Jiaying Liu, Wenhan Yang, and Zongming Guo. "Image Super-Resolution
Based on Structure-Modulated Sparse Representation", IEEE Transactions on Image
Processing (TIP), Vol.24, No.9, pp.2797-2810, May 2015.
[10] Mading Li, Jiaying Liu, Wenhan Yang, Xiaoyan Sun, and Zongming Guo. "Structure-Revealing
Low-Light Image Enhancement via Robust Retinex Model", IEEE Trans. on Image Processing
(TIP), Vol.27, No.6, pp.2828-2941, June. 2018.
[11] Fang Zhao, Jiashi Feng, Jian Zhao, Wenhan Yang, and Shuicheng Yan. "Robust LSTM-
Autoencoders for Face De-Occlusion in the Wild", IEEE Trans. on Image Processing (TIP),
Vol.27, No.2, pp.778-790, Feb. 2018.
[12] Guo-Sen Xie, Xu-Yao Zhang, Wenhan Yang, Mingliang Xu, Shuicheng Yan, and Cheng-Lin Liu.
"LG-CNN: From local parts to global discrimination for fine-grained recognition", Pattern
Recognition (PR), Vol.71, pp.118–131, Nov. 2017.
[13] Jiaying Liu, Wenhan Yang, Sifeng Xia. "Deep Image Interpolation Based on Pyramid
Location-Aware Variation Learning", IEEE Trans. on Circuit System for Video Technology
(TCSVT), Under Review.
[14] Jiaying Liu, Wenhan Yang, Shuai Yang, Zongming Guo. "D3R-Net: Dynamic Routing Residue
Recurrent Network for Video Rain Removal", IEEE Trans. on Image Processing (TIP), Under
Review.
Conference
[1] Wenhan Yang, Robby T. Tan, Jiashi Feng, Jiaying Liu, Zongming Guo, and Shuicheng Yan.
"Joint Rain Detection and Removal from a Single Image", Proc. of IEEE Conference on
Computer Vision and Pattern Recognition (CVPR), Honolulu, Hawaii, Jul. 2017.
[2] Wenhan Yang, Shihong Deng, Yueyu Hu, Junliang Xing, and Jiaying Liu. "Real-Time Deep
Video SpaTial Resolution UpConversion SysTem (STRUCT++ Demo)", Proc. of ACM
International Conference on Multimedia (ACM MM), Demo, Silicon Valley, California, U.S.,
Oct. 2017.
[3] Wenhan Yang, Jiaying Liu, and Sifeng Xia. "Variation Learning Guided Convolutional Network
for Image Interpolation", Proc. of IEEE International Conference on Image Processing (ICIP),
Beijing, China, Sep. 2017.
[4] Wenhan Yang, Jiaying Liu, Saboya Yang, and Zongming Guo. "Image Super-Resolution via
Nonlocal Similarity and Group Structured Sparse Representation", Proc. of IEEE Visual
Communications and Image Processing (VCIP), Singapore, Singapore, Dec. 2015.
[5] Wenhan Yang, Jiaying Liu, Shuai Yang, and Zongming Guo. "Novel Autoregressive Model
Based on Adaptive Window-Extension and Patch-Geodesic Distance for Image Interpolation",
Proc. of IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP),
Brisbane, Australia, Apr. 2015.
[6] Wenhan Yang, Jiaying Liu, Mading Li, and Zongming Guo. "General Scale Interpolation Based
on Fine-Grained Isophote Model with Consistency Constraint", Proc. of IEEE International
Conference on Image Processing (ICIP), Paris, France, Oct. 2014.
[7] Wenhan Yang, Jiaying Liu and, Zongming Guo. "Spatial-Temporal Recurrent Residual
Networks for Video Super-Resolution", Proc. of International Forum of Digital TV and
Wireless Multimedia Communication (IFTC), Shanghai, China, Nov. 2017.
[8] Jiaying Liu, Wenhan Yang, Shuai Yang, and Zongming Guo. "Erase or Fill? Deep Joint
Recurrent Rain Removal and Reconstruction in Videos", Accepted by IEEE Conference on
Computer Vision and Pattern Recognition (CVPR), 2018.
[9] Shuai Yang, Jiaying Liu, Wenhan Yang, and Zongming Guo. "Context-Aware Unsupervised Text
Stylization", Accepted by ACM Multimedia (ACM MM), Seoul, Korea, Oct. 2018.
[10] Qian Rui, Robby Tan, Wenhan Yang, Jiajun Su, and Jiaying Liu. "Attentive Generative
Adversarial Network for Raindrop Removal from a Single Image", Accepted by IEEE Conference
on Computer Vision and Pattern Recognition (CVPR), June 2018.
[11] Sifeng Xia, Wenhan Yang, Yueyu Hu, Siwei Ma, and Jiaying Liu. "A Group Variational
Transformation Neural Network for Fractional Interpolation of Video Coding", Proc. of Data
Compression Conference (DCC), 2018.
[12] Yueyu Hu, Wenhan Yang, Sifeng Xia, Wen-Huang Cheng, and Jiaying Liu. "Enhanced Intra
Prediction with Recurrent Neural Network in Video Coding", Proc. of Data Compression
Conference (DCC), 2018.
[13] Mading Li, Jiaying Liu, Wenhan Yang, and Zongming Guo. "Joint Denoising and Enhancement
for Low-Light Images via Retinex Model", Proc. of International Forum of Digital TV and
Wireless Multimedia Communication (IFTC), Shanghai, China, Nov. 2017. (Best paper
award)
LOY, Chen Change
Current Position:
Associate professor School of Computer Science and Engineering, Nanyang Technological
University, Aug 2018-Present
Adjunct associate
professor
Dept. of Information Engineering, The Chinese University of Hong
Kong, Aug 2018-Present
Visiting scholar SIAT, Chinese Academy of Sciences, Dec 2014-Present
Past Employment History:
Adjunct assistant
professor
Dept. of Information Engineering, The Chinese University of Hong
Kong, Mar 2018-Aug 2018
Research assistant
professor
Dept. of Information Engineering, The Chinese University of Hong
Kong, Mar 2013-Mar 2018
Post-doctoral
researcher
Vision Semantics Limited, UK, Jan 2012-Feb 2013
Post-doctoral
researcher
Queen Mary University of London, UK, Dec 2010-Dec 2011
Academic Qualifications:
Ph.D. Computer Science, Queen Mary University of London, UK, 2010
B.Eng. Electronic Engineering (1st class honours), University Sains Malaysia,
2005
Selected Publications:
1. C. Dong, C. C. Loy, K. He, and X. Tang, "Image super-resolution using deep
convolutional networks," IEEE Transactions on Pattern Analysis and Machine Intelligence
(TPAMI), vol. 38, no. 2, pp. 295-307, 2015. [Top-10 Most Popular Articles of TPAMI, Mar
2016-Present]
2. X. Wang, K. Yu, C. Dong, and C. C. Loy, “Recovering realistic texture in image super-
resolution by spatial feature modulation,” accepted to IEEE Conference on Computer
Vision and Pattern Recognition (CVPR), June 18-22, 2018, Salt Lake City, USA.
3. K. Yu, C. Dong, L. Lin, and C. C. Loy, “Crafting a toolchain for image restoration by deep
reinforcement learning,” IEEE Conference on Computer Vision and Pattern Recognition
(CVPR), June 18-22, 2018, Salt Lake City, USA.
4. W. Wu, Y. Zhang, C. Li, C. Qian, C. C. Loy, “ReenactGAN: Learning to reenact faces via
boundary transfer,” European Conference on Computer Vision (ECCV), September 8-14,
2018, Munich, Germany.
5. X. Li, C. C. Loy, “Video object segmentation with joint re-identification and attention-
aware mask propagation,” European Conference on Computer Vision (ECCV),
September 8-14, 2018, Munich, Germany.
6. S. Zhu, S. Liu, C. C. Loy, and X. Tang, “Deep cascaded bi-network for face hallucination,”
European Conference on Computer Vision (ECCV), October 8-16, 2016, Amsterdam, The
Netherlands.
7. C. Dong‚ C. C. Loy, and X. Tang, "Accelerating the super-resolution convolutional neural
network," European Conference on Computer Vision (ECCV), October 8-16, 2016,
Amsterdam, The Netherlands.
8. T.-W. Hui, C. C. Loy, and X. Tang, “Depth map super resolution by deep multi- scale
guidance,” European Conference on Computer Vision (ECCV), October 8-16, 2016,
Amsterdam, The Netherlands.
9. C. Dong, Y. Deng, C. C. Loy, and X. Tang, “Compression artifacts reduction by a deep
convolutional network,” IEEE International Conference on Computer Vision (ICCV),
December 13-16, 2015, Santiago, Chile.
10. C. Dong, C. C. Loy, K. He, and X. Tang, “Learning a deep convolutional network for
image super-resolution,” European Conference on Computer Vision (ECCV), September
6-12, 2014, Zürich, Switzerland.
11. Z. Zhang, P. Luo, C. C. Loy, and X. Tang, “From facial expression recognition to
interpersonal relation prediction,” International Journal of Computer Vision (IJCV), vol.
126, no. 5, pp. 550–569, 2018.
12. X. Li, Z. Liu, P. Luo^, C. C. Loy, and X. Tang, “Deep learning Markov random field for
semantic segmentation,” in IEEE Transactions on Pattern Analysis and Machine
Intelligence (TPAMI), 2017.
13. Y. Deng, C. C. Loy, and X. Tang, “Image aesthetic assessment: An experimental survey,”
IEEE Signal Processing Magazine (SPM), vol. 34, no. 4, pp. 80–106, 2017
14. S. Yang*, P. Luo, C. C. Loy, and X. Tang, “Faceness-Net: Face detection through deep
facial part responses,” IEEE Transactions on Pattern Analysis and Machine Intelligence
(TPAMI), 2017.
15. C. Huang, C. C. Loy, and X. Tang, “Discriminative sparse neighbor approximation for
imbalanced learning,” IEEE Transactions on Neural Networks and Learning Systems
(TNNLS), 2017.
16. Z. Zhang, P. Luo, C. C. Loy, and X. Tang, “Learning deep representation for face
alignment with auxiliary attributes,” IEEE Transactions on Pattern Analysis and Machine
Intelligence (TPAMI), vol. 38, no. 5, pp. 918–930, 2015.
17. X. Zhan, Z. Liu, J. Yan, D. Lin, C. C. Loy, “Consensus-Driven Propagation in Massive
Unlabeled Data for Face Recognition,” European Conference on Computer Vision
(ECCV), September 8-14, 2018, Munich, Germany.
18. F. Wang, L. Chen, C. Li, S. Huang, Y. Chen, C. Qian, C. C. Loy, “The Devil of Face
Recognition is in the Noise,” European Conference on Computer Vision (ECCV),
September 8-14, 2018, Munich, Germany.
19. T.-W. Hui, X. Tang, and C. C. Loy, “LiteFlownet: A lightweight convolutional neural
network for optical flow estimation,” IEEE Conference on Computer Vision and Pattern
Recognition (CVPR), June 18-22, 2018, Salt Lake City, USA.
20. Y. Rong, K. Cao, C. Li, X. Tang, and C. C. Loy, “Pose-robust face recognition via deep
residual equivariant mapping,” IEEE Conference on Computer Vision and Pattern
Recognition (CVPR), June 18-22, 2018, Salt Lake City, USA.
21. Y. He, K. Cao, C. Li, and C. C. Loy, “Merge or not? Learning to group faces via imitation
learning,” AAAI Conference on Artificial Intelligence (AAAI), February 2-7, 2018, New
Orleans, USA.
22. C. Huang, C. C. Loy, X. Tang, “Local similarity-aware deep feature embedding,” Neural
Information Processing Systems (NIPS), 2016, Barcelona, Spain.
23. W. Shi, C. C. Loy, and X. Tang, “Deep specialized network for illuminant estimation,”
European Conference on Computer Vision (ECCV), October 8-16, 2016, Amsterdam, The
Netherlands.
24. Z. Zhang*, P. Luo, C. C. Loy, and X. Tang, “Joint face representation adaption and
clustering in videos,” European Conference on Computer Vision (ECCV), October 8-16,
2016, Amsterdam, The Netherlands.
25. S. Yang, P. Luo, C. C. Loy, and X. Tang, “WIDER FACE: A face detection benchmark,”
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 26-July 1,
2016, Las Vegas, USA.
Google Scholar:
https://scholar.google.co.uk/citations?user=559LF80AAAAJ&hl=en
Recent Awards:
1. Champion, PIRM Challenge on Perceptual Super-Resolution (Third Region), 2018
2. First Runner-up, DAVIS Challenge on Video Object Segmentation, 2018
3. First Runner-up, NTIRE 2018 Challenge on Single Image Super-Resolution, 2018
4. Outstanding reviewer, British Machine Vision Conference, 2017
5. Outstanding reviewer, IEEE Conference on Computer Vision and Pattern Recognition,
2017
6. Champion, DAVIS Challenge on Video Object Segmentation, 2017
7. First Runner-up, NTIRE 2017 Challenge on Single Image Super-Resolution, 2017
8. Best application paper honorable mention, Asian Conference on Computer Vision, 2016
9. Second runner-up in ImageNet Large Scale Visual Recognition Challenge (ILSVRC)
object classification task, 2016
10. Runner-up in ImageNet Large Scale Visual Recognition Challenge (ILSVRC) object
detection task, 2014
Services:
Associate Editor
o International Journal of Computer Vision, 2018-Present
o IET Computer Vision, 2015-Present
Guest Editor
o International Journal of Computer Vision Special Issue on “Deep Learning for Face
Analysis”, 2018
o Computer Vision and Image Understanding Special Issue on “Image and Video
Understanding in Big Data”, 2017
Conference Chair
o Area Chair: CVPR 2019, ECCV 2018, BMVC 2018
o Corporate Relations Chair: ICCV 2019
o Workshop Chair: ICONIP 2014
Recent Talks:
1. Panelist, FutureChina Global Forum, Singapore, August 2018
2. Invited Talk, APAC HPC-AI Competition, Singapore, August 2018
3. Invited Talk, CVPR Workshop on Biometrics 2018, Salt Lake City, Utah, USA, June 2018
4. Invited Talk, Facial Biometrics - Technology Outlook Session, OCBC Bank, Singapore, Apr
2018
5. Guest Lecture, The Hong Kong University of Science and Technology, Hong Kong, Apr
2018
6. Keynote, SupercomputingAsia, Singapore, Mar 2018
7. Lecture, Winter School on Biometrics, Shenzhen, China, Feb 2018
