About Me

I am a Master's student in Computer Science at the University of Southern California. Prior to USC, I earned my Honours Bachelor of Science degree with High Distinction from the University of Toronto, where I double-majored in Mathematics and Computer Science. This dual background has fostered a strong foundation in math-based problem-solving and robust coding skills.

Research Interests

My research interests lie in computer graphics and vision, especially in 2D and 3D generative models, physics-based simulation, geometry processing, rendering, and 3D Gaussian Splatting. I am currently open to summer internship, research assistant and PhD opportunities that align with my passion for computer graphics and vision.

Publications

Zhenlian Miao, Guangzhu Chen, Xiaojuan Liao, Jiu Dai, Yumeng He. Self-supervised dual-layer 2D normalizing flow method for industrial anomaly detection. Applied Soft Computing (ASOC), 2024.
[Paper]

Research Experience

UCLA AIVC Lab | Visiting Student
June 2025 - Ongoing
Supervisor: Prof. Chenfanfu Jiang, Dr. Ying Jiang
Currently conducting full-time research at UCLA’s AIVC Lab on 3D articulated object generation, exploring neural and physics-guided representations for shape, joint dynamics, and scene layout—toward a first-author CVPR 2026 submission.

USC RESL Lab | Research Assistant
July 2025 - Ongoing
Supervisor: Prof. Gaurav S. Sukhatme, Dr. Jeremy Morgan
Ledding the Real2Sim pipeline for a robotic manipulation project, focusing on building a digital twin from a single RGB-D image of a cluttered tabletop scene—including object segmentation, mesh generation, and estimation of pose, scale, and placement—and importing the reconstructed scene into ManiSkill3 for simulation.

CDUT CIST | Research Assistant
May 2023 - August 2023
Supervisor: Prof. Guangzhu Chen
Faced with unlabeled industrial defects on MVTec AD and tasked to raise unsupervised detection/localization accuracy, proposed SS-DualFlow: a dual-layer 2D normalizing-flow that maps features to a Gaussian base to curb information loss and inserts an Exponential Space Attention module to focus on anomaly-salient regions.

Work Experience

Software Developer Intern | HCL Canada Inc.
August 2022 - August 2023
In a DevOps-focused internship, I streamlined build and deployment using Jenkins, built Docker images for both xLinux and pLinux, integrated SonarQube to resolve critical issues, and automated weekly log cleanups—significantly reducing human error and accelerating release cycles.

Projects

Experiement 3D Gaussian Splatting in Unreal Engine with different plugins| Unreal Engine
June 2025 - Ongoing
Experimented with 3D Gaussian Splatting in Unreal Engine with different plugins in Windows. The main objective is to compare different 3DGS plugins, including: XVERSE 3DGS, Luma AI, Akiya 3D Gaussians, Volinga UE, PFN 3DGS, and more.
[Demo Coming Soon...]

Incompressible Fluid Simulation: A Comparison | C++, OpenGL
Spring 2025
A 2D incompressible fluid simulation implemented in C++ with visualization using OpenGL and GLUT. The main objective is to compare different fluid simulation methods, including: Grid-based (Stable Fluids), Particle-based (SPH), Particle-In-Cell (PIC), hybrid PIC/FLIP method(PIC/FLIP), Affine Particle-In-Cell (APIC).
[Code] [Report] [Presentation Slides]

Inverse Kinematics with Skinning | C++
Spring 2025
Implemented skinning, forward kinematics (FK) and inverse kinematics (IK) to deform a character. The character is represented as an obj mesh.
[Code] [Demo Video]

Motion Capture Interpolation | C++
Spring 2025
This project implements motion capture interpolation techniques for an ASF/AMC mocap system. It also includes a motion comparison plotter tool that graphs selected joint angles over a specified frame range.
[Code]

Simulating a Jello Cube | C++
Spring 2025
A 3D mass-spring simulation of a jello cube, featuring structural, shear, and bend springs, collision detection, external forces, and multiple integrators (Euler, RK4, implicit).
[Code] [Project Page] [Demo Video]

Ray Tracing | C++
Fall 2024
In this assignment, we built a ray tracer. The ray tracer is be able to handle opaque surfaces with lighting and shadows.
[Code] [Project Page]

Simulating a Roller Coaster | OpenGL
Fall 2024
This assignment utilizes Catmull-Rom splines together with OpenGL's core profile shader-based lighting and texture mapping to simulate a roller coaster.
[Code] [Project Page] [Demo Video]

Height Fields Using Shaders | OpenGL
Fall 2024
This program renders a height field using OpenGL shaders, allowing interactive visualization and manipulation of terrain data derived from heightmap images. It supports various rendering modes, smooth shading, and user controls for transformation and camera adjustments.
[Code] [Project Page] [Demo Video]

Other Interesting Projects

PL-23 | Game development, Python
June 2025 - Ongoing
PL-23 is an LLM Agent-driven narrative puzzle game. We experimented with 10 different LLM Agent frameworks and prompts to develop our best-practice setup for dynamic dialogue and clue reasoning.
[Demo Coming Soon...]

Petpal | Full-Stack, Python
Fall 2023
Led a four-person team to develop a Python-based adoption platform with Django REST Framework and MySQL—implementing advanced search filters, real-time data handling, a responsive front end (HTML/CSS/JS), and secure authentication to streamline the pet adoption process.
[Code]

Courses

• USC CSCI699 - Robotic Perception - Fall 2025
• USC CSCI566 - Deep Learning and Its Applications - Fall 2025
• USC CSCI580 - 3D Graphics and Rendering - Spring 2025
• USC CSCI520 - Computer Animation and Simulation - Spring 2025 [Notes] [A1] [A2] [A3]
• USC CSCI420 - Computer Graphics - Fall 2024 [A1] [A2] [A3]
• USC CSCI570 - Analysis of Algorithms - Fall 2024

Fun Facts

• In addition to my academic pursuits, I'm also a tattoo artist. Check out my work on Instagram at @rainy.tattoo.
• I built this webpage entirely from scratch without any templates; it's a simple static site, but a great way to practice HTML and CSS.