Home             Publications             Projects             Experience             Contact

Rong Li       李 蓉  

I am a master student at SCUT. Previously, I collaborated with Anh-Quan Cao and Dr. Raoul de Charette from Inria ASTRA. I workded was an intern with Zhuangwei Zhuang of Pazhou Lab under the supervision of Prof. Dr. Mingkui Tan.

My current research is focused on 3D scene understanding, multi-sensor fusion and data efficient learning.

Google Scholar       GitHub


News

[2022-09] 1 paper is accepted to BMVC 2022.

[2022-03] We open source multi sensor embedded platform.

[2021-09] 1 paper is accepted to ICCV 2021.

Publications


Class-Prototypes for Contrastive Learning in Weakly-Supervised 3D Point Cloud Segmentation

Rong Li, Anh-Quan Cao, Raoul de Charette. BMVC, 2022
Paper Code



Perception-Aware Multi-Sensor Fusion for 3D LiDAR Semantic Segmentation

Zhuangwei Zhuang, Rong Li, Yuanqing Li, Kui Jia, Qicheng Wang, Mingkui Tan. ICCV, 2021
Paper Video Code Report




Projects


SensatUrban: Urban-Scale Point Clouds Understanding Challenge

is a BEV multi-sensor-fusion-based point cloud semantic segmentation implementation. Collaborated with Zhuangwei Zhuang.
Code



An Embedded Mobile Platform Equipped with LiDAR and Camera

is an embedded mobile platform equipped with Livox LiDAR and Camera. This platform is developed on Nvidia Jetson Xavier NX. It can be used to deploy multi-sensor based algorithms. Developed ith Zhuangwei Zhuang.
Video Code Report



Point Cloud-based Multi-Class Detection

Based on work PointRCNN, I expanded it to detect multiple classes of objects at the same time. Note: This picture is from Fast RCNN, because when this website was established, no relevant logs were found.



Human Segmentation for Intelligence Education

aims to reappear the obscured content on the blackboard by teacher to improve the quality of online courses. It based an salient object detection method. Collaborated with Zhuoman Liu, Gang Dai, Jiaqiu Zhou.
Video



Deep View Synthesis via Self-Consistent Generative Network

aims to synthesize a new view from a given input view without using explicit geometric information. Coolabrate with Zhuoman Liu and Jiaqiu Zhou.
Code Video Dataset


Experience

[2021.12 - 2022.07]      Remote Collaboration, Inria-ASTRA, Paris, French

[2020.10 - 2021.09]      Algorithm Intern, Pazhou Lab, Guangzhou, China

[2018.10 - 2019.06]      Algorithm Intern, CVTE Research, Guangzhou, China

Education

[2022]      M.S., Software Engineering, SCUT

[2019]      B.S., Software Engineering, SCUT

Honors

[2018]      National Encouragement Scholarship, China (1/43)

[2017]      National Encouragement Scholarship, China (1/43)

Contact

Email:      lirong0329@qq.com