CSE 291-J00: Topics in Image-Based Modeling and Rendering

Project Guidelines


The project for CS291-J00 is one of the most important and exciting components of the course since you will have the opportunity to develop some aspect of image-based rendering of your own. The project should not be solely theoretical in nature, but should have a strong empirical component.  The project should be at the frontier of current research, but need not necessarily move the frontier forward. Replicating the results of an innovative recent paper would be a good project, for example.  You might choose to integrate ideas from multiple papers. A number of more ambitious  term projects in the past have been significant enough to ultimately lead to conference and journal papers.  I will offer lot of latitude in selecting a project, and it can be tied into thesis research.  However, it must be distinct from the main line of your research.  That is, it must be something that you probably wouldn't have done unless you were taking this class.  

When implementing your project, you can use any language that you like. In the past, students have been particularly successful using Matlab for vision projects.  Note that the main point is to prototype one or more solutions, and not necessarily to write the fastest code.   Similarly, I do not want you to be implementing commonly available routines (e.g. matrix inversion, SVD).  You should use "off the shelf" components as much as possible so that you can tackle the interesting parts of the problem, and not get bogged down in heavy coding.  In your report, you should acknowledge the use of any code that you didn’t write yourself.

The schedule for the projects is as follows.  After each deadline, the instructor will be available to discuss the projects in person.  You should take the initiative to set up an appointment, preferably during office hours..

(1) On Thursday, February 5 at 5pm or earlier, you should hand in a project proposal that describes the goals of your project, and some initial thoughts on how you might address it.  .  This should explain explicitly and clearly what you will do.  In particular, the proposal should include:

  • a list of four to six milestones, and deadlines to achieve the milestones
  • a list of questions to be answered during the project and discussed in the report
  • experimental resources needed (e.g. digitizer, camera, software, an image database, etc).  We have a digital camera available, and I can get a hold of digital camcorder if needed, but we’ll need to plan.
  • a few recent and very closely related papers that you will build on, with full bibliographic data.

Once you pick a topic, I expect that you will stick to that topic unless you've consulted with me and written another project proposal.

The proposal should be written in well-organized continuous English, as opposed to just an outline.  Most of its text should be reusable in your final report.  In organization, proposals may resemble the project descriptions for CSE 250A, 250B, or 254.

(2) On Thursday, February 12 at 6pm or earlier, please hand in a revised project proposal.  This must follow all the instructions above, and take into account comments received from the instructor and other sources.  It should two pages long using the CVPR 2001 instructions.   You can use word, latex or another formatting package, but I highly recommend latex as it is the primary text formatting tool for publishing in computer science.  In the first phase of the project, some important tasks include:

  • formulating a clean and sensible project.
  • surveying closely related recent papers, and
  • selecting and obtaining existing software, and developing new software.

These are tasks that can and should be performed mostly in parallel, not sequentially.

(3) On Thursday, March 13 you will give a short in class presentation describing your project, including goals, methods, results, and conclusions.  More details will be provided closer to this date.

(4) On Friday, March 14 at 5pm or earlier, you should hand in a complete project report, perhaps taking into account any comments that arose during the presentation.   The report should resemble as closely as possible a good submission for a computer vision or graphics conference.  You should follow the CVPR 2001 formatting instructions and the 250A report guidelines.  This page contains some good ideas for writing technical papers, and avoid these common formatting mistakes   Hand in both hardcopy and pdf format (by e-mail). This will be graded following the CSE 250A grading criteria.  Perfect academic honesty is required.

While doing the project, remember that winning at research is similar to winning in many other fields of endeavor.

Most recently updated on Jan. 12, 2003 by D. Kriegman. Content adapted from Charles Elkan's CSE 254 Project Guidelines and Serge Belongie’s CSE 252C guidelines.