Yafei Yang | The Hong Kong Polytechnic University

Yafei Yang

I am a final year PhD student (HKPFS) in the Department of Computing at The Hong Kong Polytechnic University, supervised by Prof. Bo Yang, co-supervised by Prof. Jiannong Cao.

Previously, I got my B.Eng degree from Department of Computer Science (double majoring in Finance) at The University of Hong Kong with First Class Honors, supervised by Prof. Wenping Wang.

In my undergraduate study, I have been exchanged to UC Berkeley and National University of Singapore.

My general research interests lie in computer vision and machine learning. Currently, I am focusing on object-centric representation learning, covering the segmentation and generation of objects.

Email / Google Scholar / Github

Recent Publications

	FoundObj: Self-supervised Foundation Models as Rewards for Label-free 3D Object Segmentation Zihui Zhang, Zhixuan Sun, Yafei Yang, Jinxi Li, Jiahao Chen, Bo Yang International Conference on Machine Learning (ICML), 2026 arXiv / Code We address the challenging task of 3D object segmentation in complex scene point clouds without relying on any scene-level human annotations during training.
	PhysInOne: Visual Physics Learning and Reasoning in One Suite Siyuan Zhou, Hejun Wang, Hu Cheng, Jinxi Li, Dongsheng Wang, Junwei Jiang, Yixiao Jin, Jiayue Huang, Shiwei Mao, Shangjia Liu, Yafei Yang, Hongkang Song, Shenxing Wei, Zihui Zhang, DataTeam, Bing Wang, Zhihua Wang, Chuhang Zou, Bo Yang Conference on Computer Vision and Pattern Recognition (CVPR)*, 2026 arXiv / Code / Project Page / Data We present PhysInOne, a large-scale synthetic dataset addressing the critical scarcity of physically-grounded training data for AI systems.
	EvObj: Learning Evolving Object-centric Representations for 3D Instance Segmentation without Scene Supervision Jiahao Chen, Zihui Zhang, Yafei Yang, Jinxi Li, Shenxing Wei, Zhixuan Sun, Bo Yang Conference on Computer Vision and Pattern Recognition (CVPR), 2026 arXiv / Code We introduce EvObj for unsupervised 3D instance segmentation that bridges the geometric domain gap between synthetic pretraining data and real-world point clouds.
	RayletDF: Raylet Distance Fields for Generalizable 3D Surface Reconstruction from Point Clouds or Gaussians Shenxing Wei, Jinxi Li, Yafei Yang, Siyuan Zhou, Bo Yang International Conference on Computer Vision (ICCV), 2025 Hightlight arXiv / Code We present a generalizable method for 3D surface reconstruction from raw point clouds or pre-estimated 3D Gaussians by 3DGS from RGB images.
	unMORE: Unsupervised Multi-Object Segmentation via Center-Boundary Reasoning Yafei Yang, Zihui Zhang, Bo Yang International Conference on Machine Learning (ICML), 2025 arXiv / Code We study the challenging problem of unsupervised multi-object segmentation on single images.
	GrabS: Generative Embodied Agent for 3D Object Segmentation without Scene Supervision Zihui Zhang, Yafei Yang, Hongtao Wen, Bo Yang International Conference on Learning Representations (ICLR), 2025 Spotlight arXiv / Code We propose an unsupervised framework to separate learning objectness and search objects in 3D scenes.
	Benchmarking and Analysis of Unsupervised Object Segmentation from Real-World Single Images Yafei Yang, Bo Yang International Journal of Computer Vision (IJCV), 2024 (IF=11.6) arXiv / Springer Access / Code The journal version of our paper at NeurIPS 2022. Complete benchmark and analysis are included.
	Promising or Elusive? Unsupervised Object Segmentation from Real-world Single Images Yafei Yang, Bo Yang Advances in Neural Information Processing Systems (NeurIPS), 2022 arXiv / Project Page/ Code We systematically investigate the effectiveness of existing unsupervised models on challenging real-world images.

Teaching

Outstanding Award (top 1) for The Most Appreciated Teaching Assistant (MATA) Award 2023/24.

Recognition Award for The Most Appreciated Teaching Assistant (MATA) Award 2022/23.

Fall, 2024: Machine Learning and Data Analytics (The Hong Kong Polytechnic University).

Spring, 2024: Creative Digital Media Design (The Hong Kong Polytechnic University).

Fall, 2023: Machine Learning and Data Analytics (The Hong Kong Polytechnic University).

Spring, 2023: Creative Digital Media Design (The Hong Kong Polytechnic University).

Spring&Fall, 2021&2022: Machine Learning and Data Analytics (The Hong Kong Polytechnic University).

Spring, 2019: COMP2396 Object-oriented Programming and Java (The University of Hong Kong).

Last update: 2025.06. Thanks.