Skip to content

Hi there, I'm

Jian Shi

Computer Vision Researcher · I build

PhD Candidate @ KAUST · Co-founder of Kornia

🎓 Graduating 2026 — open to research-scientist / postdoc roles

I'm a final-year PhD candidate at KAUST (graduating 2026), advised by Prof. Peter Wonka, and a co-founder & core maintainer of Kornia — one of the most widely used open-source differentiable computer vision libraries built on PyTorch.

My research integrates geometric reasoning into generative AI for 3D perception and immersive visual computing, with a particular emphasis on stereo video synthesis — teaching machines to see, reconstruct, and synthesize the 3D world.

Portrait of Jian Shi

Updates

News

  • 2026.05 DissolveStereo accepted to ACM SIGGRAPH 2026 (Journal Track, ACM TOG).
  • 2026.05 ImmersePro accepted to ICML 2026.
  • 2026.02 Any Resolution Any Geometry accepted to CVPR 2026.
  • 2025.09 Invited talk on Agentic Computer Vision with Kornia at GOSIM Hangzhou 2025.
  • 2025.06 Two papers — VoxelKP and Amodal Depth Anything — accepted to ICCV 2025.
  • 2025.06 Received the Dean's List Award from KAUST. 🎓

Research

Selected Publications

All publications →
ImmersePro: End-to-End Stereo Video Synthesis via Implicit Disparity Learning
ICML 2026

ImmersePro: End-to-End Stereo Video Synthesis via Implicit Disparity Learning

Jian Shi, Zhenyu Li, Peter Wonka

ICML 2026

DissolveStereo: Coarse Depth Injection for Zero-Shot Stereo Video Generation
SIGGRAPH 2026

DissolveStereo: Coarse Depth Injection for Zero-Shot Stereo Video Generation

Jian Shi, Qian Wang, Zhenyu Li, Wenqing Cui, Ramzi Idoughi, Peter Wonka

ACM SIGGRAPH 2026 (Journal Track) · ACM TOG

VoxelKP: A Voxel-based Network Architecture for Human Keypoint Estimation in LiDAR Data
ICCV 2025

VoxelKP: A Voxel-based Network Architecture for Human Keypoint Estimation in LiDAR Data

Jian Shi, Peter Wonka

ICCV 2025

Dissolving is Amplifying: Towards Fine-grained Anomaly Detection
ECCV 2024

Dissolving is Amplifying: Towards Fine-grained Anomaly Detection

Jian Shi, Pengyi Zhang, Ni Zhang, Hakim Ghazzai, Peter Wonka

ECCV 2024

Differentiable Image Data Augmentation and Its Applications: A Survey
TPAMI

Differentiable Image Data Augmentation and Its Applications: A Survey

Jian Shi, Hakim Ghazzai, Yehia Massoud

IEEE TPAMI, 2024

Open Source

Open Source Leadership & Impact

Co-founder & core maintainer of Kornia · since 2020

We are actively looking for industrial and research collaborations, as well as funding, to grow Kornia and advance open-source computer vision. If you'd like to partner, sponsor, or build on top of it, I'd love to talk.

11k+ GitHub stars 3M+ monthly downloads

Activities

  • May 2026 Google Summer of Code — Org Admin & Mentor Kornia · GSoC ↗
  • May 2025 Rust for Robotics (R4R) Workshop — Kornia-rs ICRA 2025
  • May 2025 Google Summer of Code — Org Admin & Mentor Kornia · GSoC ↗
Talks & Tutorials 6
  • Sep 2025 Accessible Agentic Computer Vision with Kornia GOSIM Hangzhou 2025 · Website ↗
  • Jun 2025 ONNXSequential: Seamless ONNX Model Composition in Kornia PyTorch Day China 2025
  • Jun 2025 Bubbaloop 101: Turn Your Phone into a Security Camera in 10 Minutes Scientific Computing in Rust 2025 · Tutorial · Website ↗
  • Apr 2022 Accelerate Your Data Augmentation Pipeline with Kornia FOSSASIA Summit · YouTube ↗
  • Dec 2021 Kornia AI: Low-Level Computer Vision for AI PyTorch Developer Day 2021
  • Dec 2020 Differentiable Data Augmentation with Kornia NeurIPS 2020 Workshop · SlidesLive ↗

Background

Experience & Education

Experience

  1. 2022 — 2026

    PhD Candidate, Computer Science

    KAUST — Visual Computing Center

  2. 2020 — Present

    Co-founder & Core Maintainer

    Kornia

  3. 2021 — 2022

    Associate Researcher II

    NEC Laboratories China

  4. 2018 — 2020

    Research Assistant — Computer Vision

    The Chinese University of Hong Kong

Education

  1. 2022 — 2026

    Ph.D., Computer Science

    King Abdullah University of Science and Technology

  2. 2016 — 2018

    M.Sc. Cloud Computing — Distinction (First-Class Honours)

    University of Leicester

  3. 2011 — 2015

    B.M., Information Management and Information Systems

    Zhengzhou University of Aeronautics

Inventions

Patents

  • Issued Nov 2022 Display screen panel with dynamic electronic endoscopy graphical user interface CN307670607S
  • Filed Mar 2022 · pending Data processing method and electronic device US2023289660A1 · JP7501703B2 · CN116776137A

Get in touch

Let's talk

I'm always happy to connect about research collaborations, open-source work, or opportunities. Email is the fastest way to reach me.