Lixiang Ru

Ph.D

School of Computer Science, Wuhan University.

😅

Biography

I'm currently a computer vision researcher at Ant Group. I received my Ph.D degree from the School of Computer Science, Wuhan University, in 2023. Before that, I received my Bachelor's degree from School of Cyber Science and Engineering, Wuhan University.

My research interests include semantic segmentation, weakly-supervised semantic segmentation, self-supervised learning and other visual understanding problems.

Education

School of Computer Science, Wuhan University.

Ph.D in Computer Science,
2018.09 - 2023.06.

School of Cyber Science and Engineering, Wuhan University.

B.Sc in Information Security,
2014.09 - 2018.06.

Work Experience

Ant Group,

Senior Algorithm Engineer,
2023.07 - Now.

JD Explore Academy ,

Research Intern,
2021.03 - 2022.11.

Publications

Google Scholar

	Token Contrast for Weakly-Supervised Semantic Segmentation Lixiang Ru, Heliang Zheng, Yibing Zhan, Bo Du IEEE Conference on Computer Vision and Pattern Recognition (*CVPR*), 2023. [Paper][Code][Poster]
	Learning Affinity from Attention: End-to-End Weakly-Supervised Semantic Segmentation with Transformers Lixiang Ru, Yibing Zhan, Baosheng Yu, Bo Du IEEE Conference on Computer Vision and Pattern Recognition (*CVPR*), 2022. [Paper][Code][Poster][Project]
	Weakly-Supervised Semantic Segmentation with Visual Words Learning and Hybrid Pooling Lixiang Ru, Bo Du, Yibing Zhan, Chen Wu International Journal of Computer Vision (*IJCV*), 2022. [Paper][Code]
	Learning Visual Words for Weakly-Supervised Semantic Segmentation Lixiang Ru, Bo Du, Chen Wu International Joint Conference on Artificial Intelligence (IJCAI), 2021. [Paper][Code] [Slide] [Poster]
	An Investigation of Traffic Density Changes inside Wuhan during the COVID-19 Epidemic with GF-2 Time-Series Images Chen Wu, Yinong Guo, Haonan Guo, Jingwen Yuan, Lixiang Ru, Hongruixuan Chen, Bo Du, Liangpei Zhang International Journal of Applied Earth Observation and Geoinformation, 2021. [Paper]
	Multi-Temporal Scene Classification and Scene Change Detection with Correlation based Fusion Lixiang Ru, Bo Du, Chen Wu, IEEE Transactions on Image Processing (TIP), 2020. [Paper] [Code] [Dataset]
	Unsupervised Deep Slow Feature Analysis for Change Detection in Multi-Temporal Remote Sensing Images Bo Du, Lixiang Ru, Chen Wu, Liangpei Zhang, IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2019. [Paper] [Code]

Honors & Awards

2023.04 Outstanding Graduate Student; Wuhan University;
2022.10 National Scholarship; Wuhan University;
2022.10 Outstanding Student Scholarship; Wuhan University;
2022.05 Second Prize of Doctoral Forum; School of Computer Science, Wuhan University;
2021.11 National Scholarship; Wuhan University;
2021.11 Outstanding Student Scholarship; Wuhan University;
2020.11 Yisheng Scholarship; Wuhan University;
2020.04 IEEE GRSS Transactions Prize Paper Award; IEEE GRSS Society;
2019.11 Outstanding Student Scholarship; Wuhan University;
2016.10 Outstanding Student Scholarship; Wuhan University;
2016.10 National Encouragement Scholarship; Wuhan University;
2015.10 Outstanding Student Scholarship; Wuhan University;
2015.10 National Encouragement Scholarship; Wuhan University;

Academic Service

Journal Reviewer: TPAMI, TIP, TNNLS, TMM, TCSVT, TGRS, TAI, PR, Neurocomputing, ...
Conference Reviewer: CVPR, ICCV, ECCV, ACCV, ...

Projects

LaneDet: A model to detect and classify real-world lane lines. This project achieves > 87% lane detection accuarcy and > 93% lane classification accuarcy, and has been applied to a data annotation platform, saving about 50% annotation cost.

Demo Video


Demo Video