Can you tell me a little bit about yourself? 

Hello, I’m Chao Li , I completed my PhD in the Computer Science Department at The University of Texas at Dallas, specializing in Computer Graphics and Vision under the guidance of Professor Xiaohu Guo. Prior to earning my PhD, I received my bachelor’s degree at Beijing Jiaotong University, China, and my master’s degree at Peking University, China. After graduating from UT Dallas, I joined Reality Labs Research, Meta (formerly known as Facebook). Currently, I hold the position of Senior Research Scientist, engaged in cutting-edge research in the realms of the metaverse and augmented reality (AR) glasses.

What was your primary area of research? What type of research were you doing while you were obtaining your PhD?

3D reconstruction — it’s an intersection of Computer Graphics and Computer Vision.

Please explain your thesis in layman’s terms

In Computer Graphics and Vision, we’re trying to turn real objects into 3D digital models. This is crucial for issues like virtual reality, augmented reality and metaverse. But it’s tough when objects are moving or changing shape. Most methods need some initial hints to work, such as a rough outline or a template, and that’s not always possible.

In my thesis, I found a way to create 3D models without needing any hints, even for moving objects. I also figured out how to handle objects that change shape, and we made our methods work fast in real-time. Our techniques are better than what’s out there now.

What inspired you to pursue a PhD? 

There are two significant factors:

1. The influence of my father, who holds a PhD, played an important role. His constant encouragement inspired me to seek a PhD as a means to contribute to the advancement of human knowledge.

2. My aspiration to explore a wider range of career possibilities also drove my pursuit of a doctoral degree.

Why did you choose to pursue your PhD at UT Dallas? 

There are two primary reasons behind my choices:

1. When I applied for my PhD, I opted to do research in the realm of Computer Graphics and Vision, and it was Professor Xiaohu Guo’s research directions that particularly caught my attention.

2. A close childhood friend who also attended UT Dallas recommended the university to me based on his positive academic and living experiences here.

What did you study while at UT Dallas and what made you decide to choose what you were studying? 

I embarked on a journey to pursue a PhD in Computer Science with a specialization in Computer Graphics and Vision, a decision shaped by various factors. To start, I have held a profound interest in the domains of Computer Graphics and Vision, firmly believing that the future’s technological advancements will pivot around three-dimensional information. Furthermore, my educational background in computer science at both the undergraduate and master’s levels significantly contributed to my choice to further explore this field.

Describe your experience studying at UT Dallas.

The Comet Cruiser offers invaluable support to freshmen and proves exceptionally convenient for my daily life in the area around the university campus. The Eugene McDermott Library provides a quiet environment well-suited for focused study and research paper reading. The annual Crawfish Boil stands out as an unforgettable and delightful event. The Computer Science department’s faculty are known for their kindness and helpfulness, with special recognition for a few individuals who played pivotal roles in my academic journey. Professor Xiaohu Guo deserves special mention for his unwavering support and guidance throughout my entire PhD program. Professor Ovidiu Daescu provided invaluable one-semester research assistantship support. Furthermore, I am grateful for Shyam Karrah, whose substantial guidance greatly aided me in course selection and graduation.

Do you have any advice for future students seeking to obtain a PhD at the UT Dallas Computer Science Department? 

1. Begin by asking yourself what your career goal is: whether you would like to be a researcher in academia, industry, or a software engineer. Defining your career path and your PhD thesis topic as early as possible is crucial. 
2. Strive to have at least one paper accepted at a top-tier conference or journal within your research domain. This achievement can be a key that unlocks opportunities for internships at leading companies.
3. Embrace the age of social media; don’t hesitate to share and promote your work on various social media platforms. Doing so may lead to valuable collaboration opportunities.
4. While pursuing your academic and professional goals, don’t forget to enjoy life on the modern and beautiful campus of UT Dallas and enjoy memorable moments with your closest friends.

What type of obstacles did you overcome while obtaining your PhD? 

One of the most significant challenges I faced during my doctoral journey was identifying a research topic that warranted in-depth exploration and had the potential to make a profound impact on future technological revolutions. It took me until my 6th year of my PhD to publish my first 1^st-author paper. Fortunately, I eventually discovered the research area of “real-time 3D reconstruction of dynamic scenes,” which has emerged as a crucial field in the context of VR/AR and metaverse.

From that point forward, my research gained momentum, resulting in consecutive acceptances of two papers at ECCV (European Conference on Computer Vision), a top-tier conference in Computer Vision.

What would you say if you could go back in time to talk to yourself during the first year of your PhD? 

The advice I would give my past self is to define your PhD thesis topic and career goal as early as possible. It was my 7^th year at UT Dallas when I joined Meta as a research scientist. Upon joining Meta, I realized that had I determined my PhD thesis topic and career objectives sooner, I could have expedited my academic progress and obtained my PhD degree more promptly.

What projects did you take part in while studying at UT Dallas?  

I took part in a $2.4 million NSF research project for VR and tele-immersion , which was led by Professor Balakrishnan Prabhakaran, Professor Xiaohu Guo, Professor Mark Spong and other researchers from UC Berkeley and UIUC. In this project, one of my major contributions was leading the design and implementation of a real-time 3D reconstruction system, which captured partially-observed 3D data over time from only one single depth camera, and simultaneously reconstructed the geometry, motion, and segmentation of 3D dynamic scenes (including both human bodies and non-human objects). This system is the key to accelerating the VR/tele-immersion scene-building process by saving creators’ time and effort and enabling vivid rendering of the digital copy of everything in the VR/tele-immersion world. As a result, I published a paper at ECCV. 

Another project in which I took part was the computer-aided automated diagnosis of cleft lip and craniosynostosis, which was supported by Professor Ovidiu Daescu. It combined classic machine learning methods and the latest Convolutional Neural Network-based feature extraction. We hope more and more children will get benefits from this type of ML/DL-enabled automated diagnosis.

Where are you currently working, and how did your PhD help you get your current job? What was the job search like? 

I currently hold the position of Senior Research Scientist at Reality Labs Research, which is the VR/AR-focused research organization of Meta. 

My journey with Reality Labs Research, Meta, began with an internship, which I had the privilege of undertaking after having had a paper accepted at ECCV. During this internship, I had the opportunity to leverage my research experience and expertise in 3D reconstruction, focusing on research closely tied to real product development and making substantial contributions.

As I was searching for a full-time research position, I received multiple offers from Reality Labs Research as well as other leading tech companies. Ultimately, I decided to join my current team at Reality Labs Research, where we had a good match in research direction.

The critical thinking, problem-solving skills, and research experiences I honed during my doctoral studies were crucial in making these opportunities possible.

What type of work do you currently do? Have you kept up with your research? 

I am actively engaged in cutting-edge research in metaverse and AR glasses. Some of my recent publications include:
1. “Learning Neural Duplex Radiance Fields for Real-Time View Synthesis,” presented at the Conference on Computer Vision and Pattern Recognition (CVPR) 2023.
2. “Egocentric Activity Recognition and Localization on a 3D Map,” featured in ECCV 2022.
3. “Ego4D: Around the World in 3,000 Hours of Egocentric Video,” showcased at CVPR 2022.
My commitment to research remains steadfast. I’ve continued to explore emerging techniques, such as Neural Radiance Field (NeRF)-based 3D reconstruction, in the realm of 3D reconstruction. Furthermore, I’ve expanded my research interests to encompass other domains, including egocentric vision and computational photography.