Yang Liu (刘阳)

Email1: liuy856@mail.sysu.edu.cn Email2: aryanr@163.com

ORCID iD iconhttps://orcid.org/0000-0002-9423-9252

Biography (CV)

I’m Yang Liu (刘阳), a post-doctoral fellow at HCP Lab, School of Data and Computer Science, Sun-Yat-Sen University with co-advisor Prof. Liang Lin. Before that, I obtained my Ph.D degree of Telecommunications and Information Systems (国家一级重点学科,教育部学科评估A类学科) in 2019 at School of Telecommunications, Xidian University advised by Prof. Zhaoyang Lu. During my Ph.D study, I worked closely with Prof. Jing Li, Prof. Tao Yang and Dr. Chao Yao. Before that, I received my Bachelor Degree in telecommunication from the School of Information Engineering, Chang’an University in 2014. I am the reviewers of some top journals, including IEEE Transactions on Neural Networks and Learning Systems, IEEE Transactions on Cybernetics, IEEE Transactions on Multimedia, IEEE Transactions on Circuits and Systems for Video Technology, IEEE Signal Processing Letters, IEEE Access, Pattern Recognition, etc. My current research interests include video analysis and understanding.

Research interests

Video Analysis and Understanding, Human Action Recognition
Multi-modal Representation Learning, Knowledge Distillation
Deep Learning, Transfer Learning, Self-supervised Learning, Reinforcement Learning

News

2020-09:  I received the funds from National Natural Science Foundation of China!!!.
2020-09:  Paper Semantics-aware Adaptive Knowledge Distillation for Sensor-to-Vision Action Recognition is available.
2020-06:  I received the funds from China Postdoctoral Science Foundation (Funding rate: 19.08%).
2020-03:  I received the Fundamental Research Funds for the Central Universities.
2020-01:  My Ph.D Thesis (psw:ithd) is available now.
2020-01:  My Ph.D Thesis is published online in CNKI.
2019-10:  Codes and datasets for IEEE TIP is available.
2019-10:  One paper is accepted by IEEE Transactions on Image Processing.
2019-08:  I start working as a postdoctoral fellow at Sun-Yat-Sen University.
2019-06:  I obtain my Ph.D degree from Xidian University.

Fundings

1.  National Natural Science Foundation of China, project leader, 2021-2024
2.  General Program of China Postdoctoral Science Foundation, project leader, 2020-2021
3.  Fundamental Research Funds for the Central Universities, project leader, 2020-2022

Publications

Temporal Contrastive Graph for Self-supervised Video Representation Learning
Yang Liu, Liang Lin.
Preprint.

Semantics-aware Adaptive Knowledge Distillation for Sensor-to-Vision Action Recognition[pdf][codes]
Yang Liu, Guanbin Li, Liang Lin.
IEEE Transactions on Image Processing, under review.
(An novel end-to-end adaptive knowledge distillation framework for sensor-to-vision heterogenous action recognition problem.)

           

Cross-domain Human Action Recognition via Transfer Learning, 2019. [pdf](psw:ithd) [Defense Talk]
Yang Liu
PhD thesis, Xidian University, June 30, 2019.
Committee: Zhaoyang Lu, Jing Li, Quanxue Gao, Peiyi Shen, Guizhong Liu, Chun Qi.

           

Deep Image-to-Video Adaptation and Fusion Networks for Action Recognition [pdf][codes]
Yang Liu, Zhaoyang Lu, Jing Li, Tao Yang, Chao Yao.
IEEE Transactions on Image Processing, 2020. (CCF-A, JCR 1, IF:9.340)
(A Novel Cross-modal Adaptation Framework that utilize image features to improve the performance of video action recognition.)

           

Hierarchically Learned View-Invariant Representations for Cross View Action Recognition [pdf][codes]
Yang Liu, Zhaoyang Lu, Jing Li, Tao Yang.
IEEE Transactions on Circuits and Systems for Video Technology, 2019. (JCR 1, IF:4.133)
(A Novel Cross-view Action Recognition Framework for learning view-invariant representations.)

           

Global Temporal Representation based CNNs for Infrared Action Recognition [pdf][codes]
Yang Liu, Zhaoyang Lu, Jing Li, Tao Yang, Chao Yao.
IEEE Signal Processing Letters, 2018. (JCR 2, IF:3.268)
(A novel Optical-flow stacked difference image is proposed to capture global temporal information for infrared action recognition.)

           

Transferable Feature Representation for Visible-to-Infrared Cross-Dataset Human Action Recognition [pdf][codes][XD145 dataset]
Yang Liu, Zhaoyang Lu, Jing Li, Chao Yao, Yanzi Deng.
Complexity, 2018. (JCR 1, IF:4.621)
(The first work that uilize visible light videos as an aid to improve the infrared action recognition performance. And a visible light action dataset XD145 is built.)

           

Combining Multiple Features for Cross-Domain Face Sketch Recognition[pdf]
Yang Liu, Jing Li, Zhaoyang Lu,Tao Yang, Zijian Liu.
Chinese Conference on Biometric Recognition (CCBR), 2016.
(My first academic paper, which uses simple multi-feature fusion method to address cross-domain face sketch recognition problem.)

           

A Geometric Segmentation Method for Traffic Scenes Using Super-Pixel Label Matching
Jiangce Hu, Zhaoyang Lu, Jing Li, Yanzi Deng, Yang Liu.
Journal of Xi’an Jiaotong University, 2018.
(My co-author paper for semantic segmentation in traffic scenes.)

           

Road Scene Segmentation Using Hierarchical Graph-based Inference
Yanzi Deng, Zhaoyang Lu, Jing Li, Yang Liu.
Journal of Xi’an Jiaotong University, 2017.
(My co-author paper for semantic segmentation in traffic scenes.)

Patents (in Chinese)

[1] 卢朝阳,邓顺,李静,刘阳,许琼. “一种对装有车载监控终端的车辆监控系统及其方法”. 国家发明专利CN106600951B

Academic services

Reviewer for
IEEE Transactions on Neural Networks and Learning Systems
IEEE Transactions on Cybernetics
IEEE Transactions on Multimedia
IEEE Transactions on Circuits and Systems for Video Technology
IEEE Transactions on Human-Machine Systems
IEEE Signal Processing Letters
IEEE Access
Pattern Recognition

Selected honors & awards

2018 National Scholarship for PhD Students
2018 Excellent PhD Students
2017 First Class Academic Scholarship for PhD Students
2015 Second Class Academic Scholarship for Master Students
2013 Third Prize of National English Competition
2012 Second Prize in National Mathematical Modeling Contest