Welcome to Buyu Liu's Homepage

About Me

I am an Associate Professor at Zhejiang University from Jan. 2024, after the five-year cheerful experience at NEC Laboratories America as a senior researcher. Prior to this, I was postdoc researcher in the school of informatics at the University of Edinburgh, working with Vittorio Ferrari. I was a PhD candidate in College of Engineering and Computer Science in Australian National University from 2012 to 2016, working with Xuming He (supervisor) and Stephen Gould (advisor).

I was also working in Computer Vision Group, Canberra Research Laboratory, National ICT of Australia (NICTA) . (my previous homepage can be found here)

Research Interests

3D Scene Understanding and its Security: road layout estimation; indoor scene understanding and trajectory prediction
Semantic Video/Image Parsing
Dynamic Holistic Scene Understanding
Efficient Inference and Test-time Aware Learning
Active Learning for Structured Output

Getting involved 🙋

I am always looking for well-motivated and hard-working students who are interested in doing research with me. You can read some of my selected Publication to get a feel for the type of work that I do. I encourage students to contact me but please read the following before doing so:

To Potential PhD students: You must have good mathematical and computational skills, preferably with prior experience in machine learning and/or computer vision. Strong programming skills in Python and/or C++, and familiarity with software revision control (e.g., GitHub) is also required. When contacting me include a copy of your academic transcript and resume. A proposed research topic (as a minimum, please include (i) the research question, (ii) its relationship to previous work, and (iii) a timeline and milestones) is also preferred.
What do you expect: Why do you want to pursue a Ph.D.? This is the first question applicants need to ask themselves. Whether it's to continue your career in academia or to be more competitive in the industry, we are committed to providing assistance in four aspects: theory, programming, writing, and presentation.

Join us, and you will expect:

Weekly/Bi-weekly one-one meetings 🧑‍🏫 : discuss individual research progress.
Small-scale reading groups 📖 : share research highlights in different domains.
Research activities 🧳 : participate in top-tier conferences.
Rich team-wise activities 🏸 : if you would like to.

Note: All applications for PhD or Masters should come through the Department applications system.

Contact

buyu.zju at outlook dot com

buyu.liu at zju.edu.cn

News

We will organize ECCV24 workshop "Scalable 3D Scene Generation and 3D Geometric Scene Understanding".
We have two submissions to ACM MM24.
We have two papers accepted in CVPR 24.
We have one paper accepted in ACM MM23.
Presented our work on trajectory prediction in Berkeley Deep Drive 2023.
We have one paper accepted in CVPR23.
Served as SPC in IJCAI 2023
We have three papers accepted in CVPR22
We have two papers submitted to TIP (We rank the first on Gaze Estimation Benchmark (See either results or Within-Dataset and search for HDU_CS )! Our code can be found here)
We have one paper accepted in CVPR21
We have one paper accepted in ECCV20
...