Rui Li

Ph.D. Student

King Abdullah University of Science and Technology

Biography

My Chinese name is 李睿(Pronounce: Ray Lee). Ｉ am currently a Ph.D. student in KAUST, Computational Imaging Group, working with Prof. Wolfgang Heidrich. Before that, I obtained my bachelor’s degree in Xidian University, School of Telecommunications Engineering, and then went to University of Science and Technology of China (USTC), School of Information Science and Technology, as a master’s student.

Research: I am currently focusing on using various image measurements (e.g., light field, polarized images, RGB images, CT) to solve higher-level inverse problem or application-driven problems, such as photo-realistic 3D scene reconstruction/representation, viewpoint synthesis, segmentation, depth estimation, reflection removal, etc.

Else: I am also developing algorithmic trading system on popular cryptocurrency exchanges as well as NFT generation technology for photo-realistic scene.

Download my resumé or 简历.

Interests

Computational Photography
Photo-realistic 3D Reconstruction
Neural Rendering

Education

Ph.D. in Electrical and Computer Engineering, 2016-Present

KAUST
M.S. in Information and Communication Engineering, 2013-2016

University of Science and Technology of China
BSc in Communication Engineering, 2009-2013

Xidian University

Publications

Rui Li, Darius Rückert, Yuanhao Wang, Ramzi Idoughi, Wolfgang Heidrich (2022). Neural Adaptive Scene Tracing. In VMV, 2022.

Darius Rückert, Yuanhao Wang, Rui Li, Ramzi Idoughi, Wolfgang Heidrich (2022). NeAT: Neural Adaptive Tomography. In ACM Trans. on Graphics (Siggraph), 2022.

PDF Cite

Rui Li, Guangmin Zang, Miao Qi, Wolfgang Heidrich (2022). Shape and Reflectance Reconstruction in Uncontrolled Environments by Differentiable Rendering. https://arxiv.org/abs/2110.12975.

PDF Cite

Guangming Zang, Ramzi Idoughi, Rui Li, Peter Wonka, Wolfgang Heidrich (2021). IntraTomo: Self-supervised Learning-based Tomography via Sinogram Synthesis and Prediction. In ICCV, 2021.

PDF

Rui Li, Simeng Qiu, Guangming Zang, Wolfgang Heidrich (2020). Reflection Separation via Multi-bounce Polarization State Tracing. In ECCV, 2020.

PDF Cite Project DOI

Rui Li, Wolfgang Heidrich (2019). Hierarchical and View-invariant Light Field Segmentation by Maximizing Entropy Rate on 4D Ray Graphs. ACM Trans. on Graphics (Siggraph Asia), 2019.

PDF Cite Project DOI

Rui Li, Lu Fang (2016). Cluster Sensing Superpixel and Grouping. In CVPRW, 2016.

PDF Cite Project

Rui Li, Minjian Pang, Cong Zhao, Guyue Zhou, Lu Fang (2016). Monocular Long-term Target Following on UAVs. In CVPRW, 2016.

PDF Cite Project DOI

Projects

Neural 3D Reconstruction and Rendering

Neural rendering and differentiable rendering have become a popular and powerful toolkit for solving high-quality photorealistic 3D reconstruction and novel viewpoint synthesis. The pain points of neural rendering projects are computational efficiency and the scalability problem of solving large-scale and high-resolution cases, our project tries to improve those, to enable efficient, large-scale photorealistic 3D reconstruction, urban reconstruction, 3D intrinsic decomposition and editing.

Rui Li

XGBoost for Geo data prediction

Winner Code for 7th KAUST-NVIDIA Workshop: AI competition

Effiicient Superpixel and Grouping

Cluster Sensing Superpixel and Grouping

Light Field Segmentation, Depth Refinement

Siggraph Asia 2019 - Hierarchical and View-invariant Light Field Segmentation by Maximizing Entropy Rate on 4D Ray Graphs

Reflection Separation and Removal

ECCV 2020 - Reflection Separation via Multi-bounce Polarization State Tracing

UAVs Visual Tracking

Monocular Long-term Target Following on UAVs

Rui Li

Experience

Research Intern

SONY

Jun 2022 – Sep 2022 R&D Center, Tokyo Laboratory 23, Tokyo, Japan

Algorithm research & development in the area of computational photography, including polarization, hyper-spectral, light field and compressive sensing etc. Realization of novel functionality with the combination of unique sensor/sensing system and signal processing will be studied.

Founder

SigTrading

May 2018 – Present Internet

(1) Design real-time AI trading system that supports mainstream deep learning framework, (2)design ticker-based trading algorithm and run them in crypto exchanges, (3) design 3D NFT from real world multiview data.