Lixin Xue

I am a PhD student at ETH Zürich, supervised by Prof. Otmar Hilliges. My research interests mainly lie in computer vision and computer graphics, especially in the intersection of digital humans and neural fields.

Previously, I completed my master's degree at ETH Zürich. During my master, I did my thesis on neural faces in the Digital Humans Group at DisneyResearch|Studio, supervised by Dr. Paulo Gotardo and Dr. Derek Bradley. Prior to ETH, I did my bachelor's at Beihang University. I was lucky to work with Prof. Hao Su at UCSD and Dr. Zhirong Wu at Microsoft Research Asia.

Email  /  CV  /  Github  /  LinkedIn

profile photo
Disambiguation in Strucutre-from-Motion
Supervised by Paul‑Edouard Sarlin, Mihai Dusmanu, and Prof. Marc Pollefeys
Semester Project in Computer Vision and Geometry Group

Make COLMAP more robust to duplicated structures by filtering out wrong matches based on several state-of-the-art algorithms.

Fixed Radius Nearest Neighbor Search
Supervised by Dr. Yifan Wang, Prof. Cengiz Oztireli, and Prof. Olga Sorkine-Hornung
Semester Project in Interactive Geometry Lab
code / slides

A fixed radius nearest neighbors search implemented on CUDA with a similar interface as pytorch3d.ops.knn_points with more than an order of magnitude speedup.

Optimizations of Ball Arithmetic
Tiancheng Chen, Ran Liao, Yunxin Sun, Lixin Xue
Course Project in Advanced System Lab 2021

Design and implement an efficient library for arbitrary-precision ball arithmetic, achieving 60% of peak performance for the big integer multiplication.

Hybrid Eulerian-Langrangian Methods for Fluid Simulation
Minchao Li, Lixin Xue, Zheyu Shi
Course Project in Physically-based Simulation 2020
demos / slides

Implement and compare different hybrid eulerian-lagrangian (PIC, FLIP, APIC) methods for fluid simulation.

Road Segmentation on Aerial Images
Lixin Xue, Yu Fei, Hongjie Chen, Shengze Jin
Course Project in Computational Intelligence Lab 2020
report / code / review

Design a direction-based kernel for post-processing the network predictions to improve performance with limited data.

Sparse-to-dense Feature-metric Camera Localization
Lixin Xue, Zimeng Jiang, Le Chen
Course Project in 3D Vision 2020
report / code

Achieve better localization accuracy on multiple datasets via a feature-metric loss optimized by LM algorithm.

Ray Tracing Renderer
Minchao Li, Lixin Xue
Course Project in Computer Graphics 2019
rendering / slides

Implement a ray tracer supporting global path tracing, volumetric rendering, and advanced denoising.

Website template credits to Jon Barron