Hi, I’m Linning! I recently obtained my PhD from CUHK, Multimedia Lab (MMLab). Before that, I also had the pleasure of collaborating with brilliant minds at CUHK(SZ), working on Statistics and Operations Research. My research interests include AI for Film (🎬) and City (🏙), with a particular focus on Neural Rendering and innovative Scene Representations for large-scale spaces and daily living spaces.

I'm passionate about creating immersive spaces that highlight the natural beauty of our world and envision the spatial intelligence enriching our lives in the near future :)

✈️ Check our cool LandMark project and city-super series :)

📰 News

a box of sweet chocolates (click to expand :)

2024-02: One paper got accepted to CVPR2024🐱
2023-10: We are organizing the Third Workshop of CVEU @ICCV 2023🇫🇷
2023-08: One paper got accepted to SIGGRAPH ASIA 2023🕶
2023-07: Two papers got accepted to ICCV2023💕
2023-02 🚀 2023-06: Internship at Meta Reality Labs, Pittsburgh👓
2023-02: Two papers got accepted to CVPR2023🗽
2022-07: One paper got accepted to ECCV2022🌏
2022-03: One paper got accepted to SIGGRAPH'2022🧑🏻‍🎨
2022-01: One paper got accepted to TMM2022📹
2021-07: One paper got accepted to ICCV2021🥥
2021-06: One paper got accepted to TMM2021🍃
2020-07: Three papers got accepted to ECCV2020🐈
2020-07: One paper got accepted to MLSP2020🐰
2020-02: One paper got accepted to CVPR2020🦄
2019-07: Got admitted to the PhD program at CUHK MMLab (≧∇≦)ﾉ!!!

📚 Publications

CityNeRF: Building NeRF at City Scale (Previous version)

Yuanbo Xiangli*, Linning Xu*, Xingang Pan, Nanxuan Zhao, Anyi Rao, Christian Theobalt, Bo Dai, and Dahua Lin

ECCV 2022

Professional Experience

Research Scientist Intern, Meta Reality Lab, Pittsburgh, 2023,2024
AI for Creative Video Editing and Understanding @CVPR2024, ICCV2023, ECCV2022, ICCV2021

Education

Ph.D. in Information Engineering at The Chinese University of Hong Kong, 2020-2024
B.S. in Statistics at The Chinese University of Hong Kong, Shenzhen, 2016-2020
- GPA: 3.96/4.00, Rank: 1/152

Visiting Experience

Visiting Program at The Chinese University of Hong Kong, 2019
Summer Research at Stanford University, 2018
Summer School at University College London, 2017

Honors and Awards

Hong Kong PhD Fellowship Scheme (HKPFS)
China National Scholarships (0.2%), CHINA
Academic Performace Scholarship, Class A, CUHK(SZ)
Dean's List Award, CUHK(SZ)
Undergraduate Research Award, CUHK(SZ)

Teaching

General Physics (PHY1001)
Electronic Circuit Design (IERG1810)
Communication Labs (IERG3820)

Coursework

Click to expand

Major Courses
- Image Processing and Computer Vision
- Computer Graphics
- Distributed and Parallel Computing
- Computer Networks
- Deep Learning
- Reinforcement Learning
- Probablistic Graphic Models
- Information Theory
- Probability Theory
- Stochastic Processes
- Optimization Theory and Algorithms
- Data and Decision Analytics
- Data Mining
- Time Series
- Matrix Analysis and Computations
- Statistics and Probability
- Nonparametric Statistics
- Statistical Modelling in Financial Markets
General Courses
- Architecture
- Anatomy and Developmental Biology
- Literary London
- Superhero Movies and Comics
- Cantonese
- In Dialogue with Humanity / Nature
- Modern Chinese History and Culture

🦄 Linning Xu

📰 News

📚 Publications

(🏙) FlashGS: Efficient 3D Gaussian Splatting for Large-scale and High-resolution Rendering

(🏙) Horizon-GS: Unified 3D Gaussian Splatting for Large-Scale Aerial-to-Ground Scenes

(🏙) Proc-GS: Procedural Building Generation for City Assembly with 3D Gaussians

(🏙) Director3D: Real-world Camera Trajectory and 3D Scene Generation from Text

(🏙) GSDF: 3DGS Meets SDF for Improved Rendering and Reconstruction

(🏙) Fisheye-GS: Lightweight and Extensible Gaussian Splatting Module for Fisheye Cameras

(🏙) Octree-GS: Towards Consistent Real-time Rendering with LOD-Structured 3D Gaussians

(🏙) Scaffold-GS: Structured 3D Gaussians for View-Adaptive Rendering

(🏡) VR-NeRF: High-Fidelity Virtualized Walkable Spaces

(🏙) MatrixCity: A Large-scale City Dataset for City-scale Neural Rendering and Beyond

(🏙) AssetField: Assets Mining and Reconfiguration in Ground Feature Plane Representation

(🏙) Grid-guided Neural Radiance Fields for Large Urban Scenes

(🏙) OmniCity: Omnipotent city understanding with multi-level and multi-view images

(🎬) Dynamic Storyboard Generation in an Engine-based Virtual Environment for Video Production

(🏙) BungeeNeRF: Progressive Neural Radiance Field for Extreme Multi-scale Scene Rendering

CityNeRF: Building NeRF at City Scale (Previous version)

(🎬) Shoot360: Normal View Video Creation from City Panorama Footage

(🎬) A Coarse-to-Fine Framework for Automatic Video Unscreen

(🏙) BlockPlanner: City Block Generation with Vectorized Graph Representation

(🎬) Jointly Learning the Attributes and Composition of Shots for Boundary Detection in Videos

(🎬) A Unified Framework for Shot Type Classification Based on Subject Centric Lens

(🎬) Online Multi-modal Person Search in Videos

(🧑‍🤝‍🧑) Learn to Propagate Reliably on Noisy Affinity Graphs

(🎬) A Local-to-Global Approach to Multi-modal Movie Scene Segmentation

(📈) Exact O(n^2) Hyper-Parameter Optimization for Gaussian Process Regression