Yuehao Wang

About

I am a senior undergraduate student majoring in Computer Science at ShanghaiTech University. My topics of interest include computer graphics, computer vision and deep learning. Recently, I work on projects in neural rendering, 3D vision, and graph learning. Outside of academics, my hobbies include creating games, designing webpages, writing songs, and cycling.
mail Email: wangyh3 [at] shanghaitech [dot] edu [dot] cn
account_box [Resume] inbox [Github] camera [Photos] person_add [LinkedIn]

Research

SoGCN: Second-Order Graph Convolutional Networks

Peihao Wang*, Yuehao Wang*, Hua Lin, Jianbo Shi
Submitted to ICLR 2021, under review

We introduce a second-order graph convolution (SoGC), a maximally localized kernel, that can express a polynomial spectral filter of order K with arbitrary coefficients. We argue that the SoGC is a simple design capable of forming the basic building block of graph convolution, playing the same role as 3x3 kernels in CNNs.
Multi-view Neural Human Rendering

Minye Wu, Yuehao Wang, Qiang Hu, Jingyi Yu
Accepted by CVPR 2020

We present an end-to-end Neural Human Renderer (NHR) for dynamic human captures under the multi-view setting. NHR adopts PointNet++ for feature extraction (FE) to enable robust 3D correspondence matching on low quality, dynamic 3D reconstructions.

Course Projects

Let's CG Star
This project comprises several essential algorithms and applications of rendering and geometry processing in computer graphics domain, including a tiny encapsulation of OpenGL, Loop subdivision algorithm, ray tracing, global illumination and volume rendering.
Reinforcement Cache

Peihao Wang*, Yuehao Wang*, Rui Wang*
Excellent Course Project at CS181 Artificial Intelligence

We adopt a reinforcement learning-based method to cache replacement strategy, aiming to improve the miss rate of existing traditional cache replacement policies. The main idea of modeling is to regard the strategy as a MDP so that we can employ DRL to learn cache eviction patterns.
Offer Pool

Yuehao Wang, Zhenghao Ni, Lianghao Zhao
Excellent Course Project at CS190B Web and Text Mining

We utilize data mining techniques to predict admission of target universities. With text data crawling from related websites, we train a regression model which receives your major, TOEFL, GRE, GPA, target universities, etc., and outputs probabilities that target universities will give you offers.
Shadow Scent (Cubot) Star

Yuehao Wang, Yincen Xia, Yintao Xu, Peihao Wang, Hanbo Xiao
The Best Course Project at Roadshow of SEMI1001 Design Thinking

A mobile game prototype which is friendly to visually impaired people. After desk research and field interviews on entertainment of visually impaired people, we design this game prototype aiming to improve video game's user experience and sociability for those vulnerable people.

Teaching

  • MATH1112 Linear Algebra - Fall 2019, Fall 2020 (TA)
  • UTech Academy AI Camp - Summer 2020 (TA of Data Science Track)
  • SI100 Introduction to Computer Programming - Fall 2018 (TA)

Libraries and Tools

Pylash Engine Star
Pylash is a 2D game framework in Python, which was developed when I was a high school student. Inspired from my experience designing and developing video games, this framework is integrated with various essential modules including 2D graphics rendering (based on PySide2), event systems, tween animation, multimedia systems, SAT collision detection, etc.
Greeting Card Star
You may want to write some warm words to your best friends, parents or lover. Plain text may bore them. Greeting Card provides a magnificent animation to show the message brick by brick. Receivers just need to tap the screen until they make sense of what you want to express, maybe it's your gratitude or your love.

Entertainments

Apollo in the Sky
As a music amateur, Yuehao Wang made this toy music album with Apple's Garageband™. The album is named after 1970's United States human spaceflight program "Apollo", standing for authors' exploring in musical world is exactly like a spaceship travelling over the galaxy. Songs in this album are incorporated with various musical styles including rock and roll, Chinese traditional style, house, electro funk, R&B, etc.