Zhenxing Mi

I am a Ph.D. candidate in Prof. Dan Xu's Vision Group, at the HKUST.

I received my Bachelor's Degree and Master's Degree in Automation from AIA at HUST in China, advised by Prof. Wenbing Tao. I was an intern at Snap Research in 2024, working with Kuan-Chieh Wang, Guocheng Qian and Kfir Aberman.

I have broad research interests on 3D reconstruction such as scalable NeRF with Mixture of Experts (MoE), Multi-view Stereo and 3D surface reconstruction. I am currently interested in multimodal generation, vision-language models and 3D world models. I mainly research on scalable NeRF, multimodal generation and 3D world models currently.

Email1 / Email2 / CV / Google Scholar / Github

News

I am actively looking for summer internship opportunities in 2025, woking on 2D / 3D multimodal generation / perception powered by VLM / LLM. Please feel free to contact me if you are interested in my research.

Feb. 18, 2025: ThinkDiff posted on arXiv. Thanks for Prof. Dan Xu and friends in our Lab and Snap for their great help.

Nov. 18, 2024: LeC²O-NeRF posted on arXiv. Thanks for Prof. Dan Xu for his great help.

July. 7, 2024: Start my internship at Snap research, working with Kuan-Chieh Wang, Guocheng Qian and Kfir Aberman.

Jan. 21, 2023: One paper accepted by ICLR2023. Thanks for Prof. Dan Xu for his great help.

Aug. 13, 2022: Extension of SSRNet in CVPR2020 is accepted by TVCG.

Mar. 3, 2022: One paper is accepted by CVPR2022.

Mar. 3, 2022: New website launches.

Research

	I Think, Therefore I Diffuse: Enabling Multimodal In-Context Reasoning in Diffusion Models Zhenxing Mi, Kuan-Chieh Wang, Guocheng Qian, Hanrong Ye, Runtao Liu, Sergey Tulyakov, Kfir Aberman, Dan Xu arXiv, 2025 project page & demo / arXiv / Huggingface paper / code
	LeC²O-NeRF: Learning Continuous and Compact Large-Scale Occupancy for Urban Scenes Zhenxing Mi, Dan Xu arXiv, 2024 arXiv
	Switch-NeRF: Learning Scene Decomposition with Mixture of Experts for Large-scale Neural Radiance Fields Zhenxing Mi, Dan Xu ICLR, 2023 project page & demo / openreview / code
	Generalized Binary Search Network for Highly-Efficient Multi-View Stereo Zhenxing Mi, Di Chang, Dan Xu CVPR, 2022 project page / arxiv / code
	DeepDT: Learning Geometry From Delaunay Triangulation for Surface Reconstruction Yiming Luo, Zhenxing Mi, Wenbing Tao (Equal contribution) AAAI*, 2021 pdf
	SSRNet: Scalable 3D Surface Reconstruction Network Zhenxing Mi, Yiming Luo, Wenbing Tao (Equal contribution) CVPR, 2020 pdf SSRNet: Scalable 3D Surface Reconstruction Network Ganzhangqin Yuan, Qiancheng Fu, Zhenxing Mi, Yiming Luo, Wenbing Tao (Equal contribution) TVCG, 2022 journal

Teaching

Teaching Assistant, Introduction to Computer Science - Fall 2022/23

Teaching Assistant, Parallel Programming - Spring 2021/22

Teaching Assistant, Discrete Mathematical Tools for Computer Science - Fall 2021/22

Source code from Jon Barron, Aljaž Božič, XINTAO and Ekun.