Zhu, Runsong 朱润松

Zhu, Runsong
PhD Student

CSE, CUHK

Email: zhurunsong AT gmail DOT com
[CV]

About Me

Hi, I am Runsong Zhu (朱润松), a fourth-year PhD student in CSE, the Chinese University of Hong Kong, supervised by Prof. Chi-Wing Fu, Philip and Prof. Pheng-Ann Heng. Before that, I received my master's degree in LIESMARS, Wuhan University supervised by Prof. Bisheng Yang and Prof. Zhen Dong in 2022 and bachelor's degree in Central South University in 2019.

Research interests: In general, I am interested in spatial intelligence for the physical world. My research initially focused on 3D geometric analysis, including normal estimation and surface reconstruction. Later, I explored open-world 3D scene understanding toward building spatial foundation models, such as unified scene reconstruction and understanding. Currently, my ongoing projects lie at the intersection of scene understanding, reconstruction, and generation. In addition, I have also contributed to projects on 3D generation and assembly.

News

[Apr. 2026] EPS3D is accepted to ICML 2026.
[Sept. 2025] COS3D is accepted to NeurIPS 2025.
[Jul. 2025] Unified-Lift is accepted to CVPR 2025.
[Jul. 2024] PCF-Lift is accepted to ECCV 2024.
[Oct. 2023] SSP is accepted to WACV 2024.
[Aug. 2022] Started PhD journey in CSE, the Chinese University of Hong Kong
[Jul. 2021] AdaFit was accepted to ICCV 2021 (oral)!
[Jun. 2021] joined Tencent AI-Lab as a research intern
[Sept. 2019] Started master journey LIESMARS, Wuhan Univeristy

First-Author Publications

| |

	EPS3D: End-to-End Feed-Forward 3D Panoptic Segmentation Runsong Zhu, Jiaxin Guo, Xiaoyang Guo^†, Zhengzhe Liu^†, Ka-Hei Hui, Wei Yin, Kai Chen, Wei Chen, Weiqiang Ren, Yunhui Liu, Pheng-Ann Heng, Chi-Wing Fu. (^† corresponding authors) International Conference on Machine Learning (ICML), 2026. [Code][Paper] [Poster]
	COS3D: Collaborative Open-Vocabulary 3D Segmentation Runsong Zhu, Ka-Hei Hui, Zhengzhe Liu, Qianyi Wu, Weiliang Tang, Shi Qiu, Pheng-Ann Heng, Chi-Wing Fu. The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025. [Code][Paper] [Poster]
	Unified-Lift: Rethinking End-to-End 2D to 3D Scene Segmentation in Gaussian Splatting Runsong Zhu, Shi Qiu, Zhengzhe Liu, Ka-Hei Hui, Qianyi Wu, Pheng-Ann Heng, Chi-Wing Fu. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025. [Code][Paper] [Poster]
	PCF-Lift: Panoptic Lifting by Probabilistic Contrastive Fusion Runsong Zhu, Shi Qiu, Qianyi Wu, Ka-Hei Hui, Pheng-Ann Heng, Chi-Wing Fu. European Conference on Computer Vision(ECCV), 2024. [Code][Paper] [Poster]
	SSP: Semi-Signed Prioritized Neural Fitting for Surface Reconstruction From Unoriented Point Clouds Runsong Zhu, Kang Di, Ka-Hei Hui, Yue Qian, Shi Qiu, Zhen Dong, Linchao Bao, Pheng-Ann Heng, Chi-Wing Fu. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision(WACV), 2024. [Project Page][Paper] [Poster]
	AdaFit: Rethinking Learning-Based Normal Estimation on Point Clouds Runsong Zhu, Yuan Liu, Zhen Dong, Yuan Wang, Tengping Jiang, Wenping Wang, Bisheng Yang. (* equal contribution) Proceedings of the IEEE/CVF International Conference on Computer Vision(ICCV), 2021 (Oral, top 3%). [Project Page][Paper] [Poster] [Video]