Avatar

Kaixuan Wei

PhD Student @ KAUST

VCC Imaging

Biography

I am a PhD student at King Abdullah University Of Science And Technology (KAUST), under the supervision of Prof. Wolfgang Heidrich. I received B.S. of Electronic Engineering and M.S. of Computer Science from Beijing Institute of Technology in 2018 and 2021, respectively. From 2021 to 2023, I was a “Ghost/secret PhD student” at Princeton University due to the uncontrollable political issues (pp. 10043), which drastically impacted my career and life. I was a research intern at University of Cambridge and Microsoft Research Asia in 2019 and 2018, respectively.

My research interest lies at the intersection of machine learning, deep learning, optimization and statistical modeling/inference with applications for computational photography/imaging/optics and computer vision.

I hold a strong believe that computational imaging/optics will ultimately reshape everyone’s life, and I do research for fun. I have a strong built-in anti-anxiety filter in mind, and my MBTI is INTP-A.

I regularly serve as a reviewer for prestigious international conferences and journals in computer vision and machine learning. Please check my CV For details.

Interests

  • Computational Optics
  • Computational Imaging
  • Computational Photography

Education

  • PhD Student of Computer Science, 2024.1 -

    King Abdullah University of Science and Technology

  • M.S. of Computer Science, 2018 - 2021

    Beijing Institute of Technology

  • B.S. of Electronic Engineering, 2014 - 2018

    Beijing Institute of Technology

Publications

Crafting Object Detection in Very Low Light

The 32nd British Machine Vision Conference (BMVC 2021), Virtual, Online

Physics-based Noise Modeling for Extreme Low-light Photography

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2021

Tuning-free Plug-and-Play Proximal Algorithm for Inverse Imaging Problems

The 37th International Conference on Machine Learning (ICML 2020), Vienna, Austria, Outstanding Paper Award, Acceptance Rate: 0.04%

A Physics-based Noise Formation Model for Extreme Low-light Raw Denoising

The 36th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2020), Seattle, USA, Oral Presentation, Acceptance Rate: 5%

3D Quasi-Recurrent Neural Network for Hyperspectral Image Denoising

IEEE Transactions on Neural Networks and Learning System, 2020

Single Image Reflection Removal Exploiting Misaligned Training Data and Network Enhancements

The 35th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2019), Long Beach, USA