Chen Liu

Hi :) I am Chen, now a Computer Graphics Ph.D. student at University College London (UCL), working with Prof. Tobias Ritschel and Prof. Niloy Mitra.

Before that, I had a Master from UCL and obtained my Bachelor from Beihang University. During my undergrad, I interned at Prof. Xianglong Liu’s Lab and worked at ByteDance as a graphics engineer.

Talking about research aspirations, my goal is to create a world satiating our “childish sense” of grand adventures. Here is a list of topics that shape the core of my works and interests: Appearance Modelling, Photorealistic Rendering, Data-driven Image Synthesis, and many others more – AI, of course.

News

Jun 26, 2025	One paper got accepted to ICCV 2025 at Hawaii, Aloha!
Jun 01, 2025	I will visit Elliott this summer at University of Cambridge.
Mar 24, 2025	I visited and gave a talk at MAVERICK, Inria Grenoble Rhône-Alpes.
Mar 08, 2025	I will attend Eurographics 2025 in London. Happy to meet up!
Feb 26, 2025	Honored to receive The Rabin Ezra Scholarship.

Publications

Generative Video Bi-flow

Chen Liu, and Tobias Ritschel

International Conference on Computer Vision (ICCV 2025)

PDF Cite

We learn ODE flows between consecutive frame distributions, hence generating videos efficiently and robustly.
@article{liu2025generative, title = {Generative Video Bi-flow}, author = {Liu, Chen and Ritschel, Tobias}, journal = {arXiv preprint arXiv:2503.06364}, year = {2025}, }

Neural Differential Appearance Equations

Chen Liu, and Tobias Ritschel

ACM Transactions on Graphics (SIGGRAPH Asia 2024)

Accepted as TOP 5%

Details PDF Code Dataset Cite

NDAEs govern time-varying appearances, e.g., rusting and weathering, in terms of RGBs or even BRDFs.

@article{liuNeuralDifferentialAppearance2024,
  title = {Neural Differential Appearance Equations},
  author = {Liu, Chen and Ritschel, Tobias},
  journal = {ACM Transactions on Graphics},
  volume = {43},
  number = {6},
  pages = {1--17},
  year = {2024},
  doi = {10.1145/3687900},
}

Learning to Learn and Sample BRDFs

Chen Liu, Michael Fischer, and Tobias Ritschel

Computer Graphics Forum (Eurographics 2023)

Details PDF Code Cite

A set of most informative locations to sample BRDFs can be learned via meta-learning.

@article{liuLearningLearnSample2023,
  title = {Learning to Learn and Sample BRDFs},
  author = {Liu, Chen and Fischer, Michael and Ritschel, Tobias},
  journal = {Computer Graphics Forum},
  volume = {42},
  number = {2},
  pages = {201--211},
  year = {2023},
  doi = {10.1111/cgf.14754},
}

Projects

A collection of my useful code-bases
Graphics courses on Rendering / Simulation / Geometry with code
Path Tracer (Ray-Tracing-in-One-Week Series)

Services

Reviewer: SIGGRAPH Asia, Pacific Graphics, Computer Graphics Forum, CVPR, ICCV

Volunteer: EG 2025 [💼, 📸], EGSR 2024 [📀, 📸], GAMES101 Grader

Teaching Assistant at UCL