Ruian He | 何瑞安

I am a 4th-year PhD student in School of Computer Science at Fudan University, supervised by Prof. Bo Yan. Before that, I received my Bachelor's degree of Computer Science at Fudan University in 2021. My research interests lie in the field of low-level vision, with application to rendering and microscopy imaging.

Email: rahe16 (at) fudan.edu.cn.

CV  /  Google Scholar  /  Semantic Scholar  /  DBLP  /  Github  /  LinkedIn

• [March 2025] One paper accepted by Nature Biomedical Engineering 🎉

(*: Equal contribution)

We show that pretraining on approximately one million synthetic images allows a retinal foundation model trained on only 16.7% of 904,170 real images to match or exceed the performance of RETFound trained on the full dataset across nine tasks.

We introduce MDSR-Zero, a zero-shot online learning method designed for plug-and-play noise suppression and super-resolution of microscopy videos.

We proposes FacialFlowNet, a novel large-scale facial optical flow dataset, and the Decomposed Facial Flow Model, the first method capable of decomposing facial flow.

We provide a universal fluorescence microscopy-based image restoration (UniFMIR) model to address different restoration problems, and show that UniFMIR offers higher image restoration precision, better generalization and increased versatility.

We recognize the shared context and mechanisms between frame supersampling and extrapolation, save up to 75% of time against the conventional two-stage pipeline that necessitates 17ms.

Through theoretical analysis, we find the pre-trained large vision models are helpful in optical flow estimation, and SAM is suitable for solving the fragmentation problem.

We develop a Context-Aware Iteration Policy Network, which determines the optimal number of iterations per sample and reduce FLOPs by about 40%/20% for the Sintel/KITTI datasets.

We develop an Uncertainty-Guided Spatial Pruning (UGSP) architecture to skip redundant computation for efficient frame interpolation dynamically.

We propose an optical flow model, MVFlow, which uses motion vectors to improve the speed and accuracy of optical flow estimation for compressed videos.

Our method utilizes the frame difference of multiple previous frames and fuse multiple spatial-temporal scales.

We propose a novel fine-grained blind face inpainting framework and build up a new dataset called CelebO-3D.

We propose a task-specific framework which first combines occlusion discarding and feature completion together to reduce the interference of occlusions on instance level.

• Conference Reviewer / Program Committee Member for CVPR (2022, 2024, 2025), NIPS (2024), ICCV (2023), ECCV (2024), AAAI (2023, 2024, 2025), IEEEVR (2023), ACMMM (2024).
• Journal Reviewer for TOG(2024), TPAMI (2024), TCSVT (2024).
• Teaching Assistant for COMP130018.01 Computer Graphics A (2023 Spring, 2024 Spring, 2025 Spring), COMP130014.01 Compiler (2022 Autumn).

• Second place in the 3rd Anti-UAV Challenge in CVPR 2023.
• Third Prize in Fast-Motion VFI Challenge in IACC 2023.

• National Scholarship (Top 1%). 2024.
• Jinrirencai Scholarship (10 students in Fudan). 2024.
• Award for Outstanding Ph.D. Students. 2021, 2023.
• Boxue Scholarship (Top 10%). 2017.
• Award for Outstanding Undergraduate Students. 2018, 2019, 2020.