Home People Research Publications Demos
News Jobs Prospective
About Internal

Information for Prospective Students

We receive many more applications than the number of positions available at SVCL. In fact, it is not even possible to reply to most emails. This page lists some requirements for the different types of positions.


Post-doc applicants are expected to have extensive experience in computer vision and machine learning, and a track record of publications on the major vision and/or learning conferences. In particular, the candidates are expected to have published at least three papers in CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, JMLR, PAMI, or TIP. The candidates should also have experience with deep learning, and strong analytical abilities. Post-doc appointments are contingent on the availability of funds and last for one year, but can be renewed. If you think you qualify, feel free to send us an email.

Visiting scholars:

We only consider visiting scholars who are self-funded. The scholars must be doing a PhD degree or post-grdaute work in computer vision and machine learning, and are expected to have published at least one paper in CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, JMLR, PAMI, or TIP. The scholars should also have experience with deep learning, and strong analytical abilities.

Undergraduate visitors:

Because there are so many UCSD students looking for research opportunities, we have no positions for non-UCSD undergraduates. If you are not a UCSD student, please do not send email for short-term internships and other visiting positions. We will not be able to follow up. Exceptions only hold for institutions that have joint programs with UCSD. These students are considered UCSD students from our point of view. However, it is the student responsibility to be admitted into the joint program and take all the other steps necessary.

Prospective graduate students:

The first requirement to be a graduate student at SVCL is to be accepted as a graduate student by the Department of Electrical and Computer Engineering at UCSD. We have no control over the admissions process, but we can make recommendations to the admission commitee. The process is tremendously competitive, only a few percent of the applicants are admitted. The following are some steps you can take to maximize your chances:

  • apply explicitly to the programs that you want to work in. If you are interested in SVCL, specifically select the Signal and Image Processing (SIP); Intelligent Systems, Robotics, and Control (ISRC); or Machine Learning and Data Science (MLDS) Areas as the area of interest for your studies. Otherwise, the SVCL faculty will probably never see your application. In addition to this, specifically mention the faculty that you want to work with in your application. You can list multiple professors, but please be specific, e.g. list at most three faculty members in the SIP area.
  • make sure to select Faculty that match well your research interests. It is not uncommon for us to be contacted by potential students that want to work on wireless communications, multimedia networking, and things like that. Needless to say, these contacts invariably lead to a negative response. If you cannot understand the research emphasis of SVCL, SVCL is probably not the right lab/research group for you.
  • show ability to do independent research. Although this is not a mandatory requirement, we receive a fair number of applications from students that have done undergraduate research, some having even already published in the main vision and learning journals and conferences. These days, the most competitive applicants have usually at least submitted a paper to CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, JMLR, PAMI, or TIP. This does not mean that you should not apply if you have not done research yet, but the other components of your application will have to be very strong.
  • students that already have a master's degree are expected to have published at least one first-author paper in CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, JMLR, PAMI, or TIP. There can be exceptions to this rule, e.g. a student that has submitted a paper currently under review, or even a student that has not submitted any paper but has done strong work in these areas and is expected to submit very soon. But these exceptions tend to be few.
  • pay close attention to your resume. This should include all graduation dates and GPAs, and include a list of publications, if you have one. The resume is the first thing that our readers look at at and plays an important role in their evaluatoins. Beyond this, you should mention the most important research efforts and the resulting papers in your statement of purpose and have your references discuss them.
  • in your statement of purpose, try to be as specific as possible about the projects within SVCL that you find interesting, and which you would like to be involved in. Note that there are usually projects for which a web page is not available yet. It is a good idea to check the publications section to find out about the latest projects. If possible, briefly describe how you would attack these problems.
UCSD Graduate students:

If you are a graduate student at UCSD, you should still read the above list of requisites and see how well you meet them. To become a member of SVCL, you need to fulfill one additional requirement, which is to have taken the courses which cover the basic areas of SVCL research. Currently these are the ECE271A-C course series: two courses on statistical learning and one on deep learning. 271A is mandatory. The remaining are optional for you to be considered by the lab, but mandatory for all lab members. While we may admit students that have not fulfilled this requirement, this only happens in exceptional circumstances. Note that 271A is only offered in the Fall, so you need to plan accordingly.

UCSD Undergraduates:

UCSD Undergraduates can perform internships at SVCL. This usually involves working with a graduate student, on an interesting research problem. It is unlikely that we will have funding for undergraduate research, but the ECE Department frequently has programs that fund internships. Please inquire with the department. Otherwise, undergraduates can work for credit, and we tend to have students that stay at SVCL over the summer. If you are a member of a minority group, you may consider applying for programs such as STARS or UCLeads, that can fund your research at SVCL. This applies to both UCSD and non-UCSD students, but only students from the US.