"Luck is what happens when preparation meets opportunity."


  Yongcheng Jing

    M.Sc. student in College of Computer Science and Technology

    Zhejiang University

    E-mail: ycjing AT zju DOT edu DOT cn

Short Bio [CV]

I am currently a graduate student in Microsoft Visual Perception Laboratory (VIPA) of Zhejiang University and Alibaba-Zhejiang University Joint Research Institute of Frontier Technologies (AZFT). I am fortunately advised by Prof. Mingli Song and Prof. Yizhou Yu.

I am also very fortunate to collaborate with Prof. Yezhou Yang and Prof. Dacheng Tao.

Before coming to Zhejiang University, I received the B.E. degree from the Elite Class of Computer Science at South China University of Technology. I was working on pattern recognition and computer vision under the supervision of Prof. Patrick P. K. Chan. I was awarded the Early Graduation Honor and graduated one year ahead of the expected graduation date in 2016. After that I became an M.Sc. student at Zhejiang University with the entrance exam free.

Research Interests

  • Texture Modelling (M.Sc.)

  • Human Parsing and Fashion Outfit (M.Sc.)

  • Image Aesthetic Assessment (M.Sc.)

  • Biometric-based Authentication (Bachelor)


2016.9 - present, Computer Science and Technology, Zhejiang University (M.Sc. with Entrance Exam Free)

2013.9 - 2016.7,  Computer Science and Technology, South China University of Technology (Bachelor with Early Graduation Honor)

Selected Projects & Publications

[Google Scholar]

Stroke Controllable Fast Style Transfer with Adaptive Receptive Fields
Y Jing, Y Liu, Y Yang, Z Feng, Y Yu, D Tao, M Song.
European Conference on Computer Vision (ECCV), 2018.
[arXiv] [Project Page] [Code] [Video] [Supplementary] [Blog (in Chinese)]

The Fast Style Transfer methods have been recently proposed to transfer a photograph to an artistic style in real-time. This task involves controlling the stroke size in the stylized results, which remains an open challenge. In this paper, we present a stroke controllable style transfer network that can achieve continuous and spatial stroke size control.

Neural Style Transfer: A Review
Y Jing, Y Yang, Z Feng, J Ye, Y Yu, M Song.
arXiv preprint, 2017
[arXiv] [Project Page] [Supplementary] [Blog (in Chinese)] [Media Coverage]

Neural Style Transfer is an important computer vision task which is to exploit Convolutional Neural Networks to migrate the semantic content of one image to different styles. This review aims to present and summarize recent progress towards Neural Style Transfer, as well as discussing its various applications and open problems for future research.

Finer-Net: Cascaded Human Parsing with Hierarchical Granularity
J Ye, Z Feng, Y Jing, M Song.
IEEE International Conference on Multimedia and Expo (ICME), 2018. (Oral)

In this paper, we propose a cascaded human parsing architecture with hierarchical granularity from coarse to fine. The proposed framework can be extended according to the demands of concrete segmentation precision. Furthermore, we effectively use the human pose information predicted by convolutional pose machine beforehand to further improve the final results.

Interpretable Partitioned Embedding for Customized Fashion Outfit Composition
Z Feng, Z Yu, Y Yang, Y Jing, J Jiang, M Song.
ACM International Conference on Multimedia Retrieval (ICMR), 2018. (Oral)
[PDF] [Media Coverage]

In this paper, we propose a partitioned embedding network to learn interpretable representations from clothing items. Our proposed network can extract interpretable embeddings of fashion outfit items. Based on the extracted embeddings, we put forward a weakly-supervised fashion outfit composition model which depends solely on a large number of outfits without quality scores. We also propose a customized fashion outfit composition scheme.

Graph-based Color Gamut Mapping Using Neighbor Metric
Z Feng, Y Jing, C Zhang, R Xu, J Lei, M Song.
IEEE International Conference on Multimedia and Expo (ICME), 2017. (Oral)

In this paper, a new Multi-source Shortest Paths Algorithm (MSSPA) is proposed to establish color mapping relationships between out-of-gamut colors and colors in gamut boundary. Experimental results show that our method achieves superior performance on the aspect of keeping accuracy and preserving details compared with HPMinDE and SGCK.


A Review on Adversarial Attacks (June 4, 2018) [Slides]

Deep Image Aesthetics (Jan 17, 2018) [Slides]

Fast Neural Style Transfer (Aug 14, 2017) [Slides]

Recent Advances in Image Reconstruction (July 17, 2017) [Slides]

Generative Adversarial Networks: Recent Advances and Popular Application (Mar 20, 2017) [Slides]

Visualizing Convolutional Neural Networks: Image Reconstruction (Oct 17, 2016) [Slides]

Honors & Awards

Outstanding Reviewer for JVCI, 2017

National Scholarship in ZJU, 2017

Graduate of Merit/Triple A graduate in ZJU, 2017

Award of Honor for Graduate in ZJU, 2017

Academic Scholarship in ZJU, 2016

Early Graduation Honor (top 5%) in SCUT, 2016

“Hong Ping Chang Qing” Academic Scholarship, 2016

Honorable Mention in Interdisciplinary Contest In Modeling (administered by COMAP), 2015

Volunteer Stars Awards in SCUT, 2015

First-class and Third-class Student Scholarship in SCUT, 2013 - 2016

The Elite Class of Computer Science (40 best students are chosen from about 1500 freshmen in SCUT), 2013 - 2016

Academic Activities

Journal Reviewer:

  • Information Sciences (INS)

  • Journal of Visual Communication and Image Representation (JVCI)

  • Journal of Electronic Imaging (JEI)

Working Experience

  • Algorithm Engineer Intern, Taobao AI Team, Alibaba Group