Zhenyu Li 李震宇

Phd Student

Expected Graduation: 2026.09. Open to research scientist / engineer roles
King Abdullah University of Science and Technology (KAUST)

Ex: Meta, ByteDance, DiDi, SenseTime;
Email: zhenyu.li.9955@gmail.com;
Github: https://github.com/zhyever
Google Scholar: Google Scholar Link
CV: CV Link

Biography

I'm a final-year PhD student at King Abdullah University of Science and Technology (KAUST), advised by Prof. Peter Wonka.

I got my B.E. and M.S degrees in computer science at Harbin Institute of Technology, Harbin, China.

My research has been centered on 3D reconstruction, and I have recently developed a strong interest in other domains, including generative models, RL, VLM, and LLM.

News

2026.05: CriterAlign preprint released on arXiv, criterion-centric LLM-as-a-judge for code preference prediction.
2026.02: Two papers, EgoPoseFormer V2 and URGT, accepted to CVPR 2026! EgoPoseFormer V2 was completed during my internship at Meta. It's selected as a highlight paper. I sincerely thank all my peers and mentors for their guidance and support.
2026.01: Two papers, PatchRefiner V2 and DepthAnything3, accepted to ICLR 2026! Glad and honored to contribute to DepthAnything3, which is selected for an oral presentation.
2025.08: I started my internship at Meta Reality Labs, Zurich, Switzerland!
2025.06: One paper, Amodal Depth Anything, accepted to ICCV 2025!
2025.01: I started my internship at ByteDance Seed, Beijing, China!
2024.07: One paper, PatchRefiner, accepted to ECCV 2024!
2024.03: One paper, PatchFusion, accepted to CVPR 2024!

Experience

2025.08 - 2026.02, Research Intern (Body Tracking for AR/VR), Meta Reality Labs, Zurich, Switzerland
2025.01 - 2025.07, Research Intern (3D Reconstruction) (TopSeed Candidate), ByteDance Seed, Beijing, China
2022.08 - 2023.04, Research Intern (ADAS) (Elite Camp), DiDi Cargo, Beijing, China
2022.01 - 2022.07, Research Intern (ADAS), SenseTime, Shanghai, China
2021.03 - 2021.09, Development and Research Intern (ADAS), SenseTime, Shanghai, China

Awards

Dean’s List Award 2024, award to top students by KAUST CEMSE for their academic achievements.
1^st place at VCL 2023 Challenge, Multitask Learning for Robustness Track! (ICCV 2023 Workshop)
China National Scholarship 2022.
3^rd place at SSLAD 2022 Challenge, 3D Object Detection Track! (ECCV 2022 Workshop)
2^nd place at Mobile AI&AIM 2022 Challenge, Monocular Depth Estimation Track! (ECCV 2022 Workshop)

Codebase

	Monocular Depth Estimation Toolbox Zhenyu Li 2022 [Code] [Bibtex]
@misc{lidepthtoolbox2022, title={Monocular Depth Estimation Toolbox}, author={Zhenyu Li}, howpublished = {\url{https://github.com/zhyever/Monocular-Depth-Estimation-Toolbox}}, year={2022} }

Selected Publications

	Artic-O: End-to-End Articulated Object Reconstruction via Latent Geometry Learning Xuyang Wang, Zhenyu Li, Jian Ding, Habib Slim, Peter Wonka, Hongdong Li, Mohamed Elhoseiny Arxiv [PDF]
	CriterAlign: Criterion-Centric Rationale Alignment for Code Preference Judging Zhenyu Li, Aleksandar Cvejic, Zehui Chen, Peter Wonka Arxiv [Project Page] [PDF]
	DivRL: Disentangled Self-Similarity Rewards for Diverse Subject-Driven Generation Qian Wang, Zhenyu Li, Abdelrahman Eldesokey, Peter Wonka ECCV 2026 [Project Page] [PDF]
	EgoPoseFormer v2: Accurate Egocentric Human Motion Estimation for AR/VR Zhenyu Li, Sai Kumar Dwivedi, Filip Maric, Carlos Chacon, Nadine Bertsch, Filippo Arcadu, Tomas Hodan, Michael Ramamonjisoa, Peter Wonka, Amy Zhao, Robin Kips, Cem Keskin, Anastasia Tkach, Chenhongyi Yang CVPR, 2026 (Highlight) [Project Page] [PDF]
	Any Resolution Any Geometry: From Multi-View To Multi-Patch Wenqing Cui, Zhenyu Li, Mykola Lavreniuk, Jian Shi, Ramzi Idoughi, Xiangjun Tang, Peter Wonka CVPR, 2026 [Project Page] [PDF] [Code]
	Depth Anything 3: Recovering the Visual Space from Any Views Haotong Lin, Sili Chen, Hao Liew Jun, Y. Chen Donny, Zhenyu Li, Guang Shi, Jiashi Feng, Bingyi Kang ICLR, 2026 (Oral) [Project Page] [PDF] [Code]
	Amodal Depth Anything: Amodal Depth Estimation in the Wild Zhenyu Li, Mykola Lavreniuk, Shariq Farooq Bhat, Peter Wonka ICCV, 2025 [Project Page] [PDF] [Code]
	PatchFusion: An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation Zhenyu Li, Shariq Farooq Bhat, Peter Wonka CVPR, 2024 [Project Page] [PDF] [Code]
	AutoAlignV2: Deformable Feature Aggregation for Dynamic Multi-Modal 3D Object Detection Zehui Chen, Zhenyu Li, Shiquan Zhang, Liangji Fang, Qinhong Jiang, Feng Zhao ECCV, 2022 [PDF] [Code]
	BinsFormer: Revisiting Adaptive Bins for Monocular Depth Estimation Zhenyu Li, Xuyang Wang, Xianming Liu, Junjun Jiang Ranked 1^st on KITTI depth estimation benchmark (Feb, 2022). Transactions on Image Processing [PDF] [Code]

Service

Conference Reviewer: CVPR, ICCV, ECCV, ICML, ICLR, SIGGRAPH.