|
MIT CSAIL
6.819/6.869: Advances in Computer Vision |
|
Fall 2017 |
|
Project Topics
Final Project is an opportunity for you to apply what you have learned in class to a problem of your interest in computer vision. We recommend you work in teams. Each team can be up to 2 people.
Option 1: Choose from the suggested topics.
You could choose from the list of suggested topics (updating):
Option 2: Your own project.
Proposal due: Thu Oct 26. Upload to stellar.
You could select a topic in computer vision that interests you most and work on it as your course project. Potential projects could be based on applications and models:
- Applications: You would apply the techniques of computer vision to some specific applications with your background and interest, such as some image processing mobile APP and video recognition software.
- Models: You would build up some new models, or improve previous models or methods, then evaluate the proposed models systematically on some standard image datasets to show the improvement.
Please clearly specify and justify: (1)what will be the approach; (2)why is it interesting; (3) how will you evaluate success.
You could take a look at the Resources (image datasets and papers) in the
Course Materials for some inspiration. Before proceeding this option, please find teammates through Piazza then draft
a summary of the project proposal together and send it to the instructors for plausibility analysis, then set up a meeting about the project detail.
Forming Groups
The rules for forming groups are:
- 6.819 and 6.869 students can mix up and form a team; these teams will be evaluated by 6.869 standards. Please specify 6.819 or 6.869 for each team member in the report.
- Groups can have up to 2 members.
- Reports should be individually submitted and it should highlight the contributions of each team member on a section of the paper.
Report
The report should be 4 pages for 6.819, and 6 pages for 6.869 (the upper limit of 6 pages is strict!), including references in CVPR format. It should be structured like a research paper, with sections for Introduction, related work, the approach/algorithm, experimental results, conclusions and references. Project reports should be individually submitted and the contributions of each team member should be clearly described.
Regarding the reports:
- Each student should submit an individual copy. All the members of the group can share figures and text. But each copy should have one section that will be individual and should describe the contribution made by the student.
- The rest of the document can be identical across members.
- Each copy should include the names of all the collaborators.
You should describe and evaluate what you did in your project, which may not necessarily be what you hoped to do originally. A small result described and evaluated well will earn more credit than an ambitious result where no aspect was done well. Be accurate in describing the problem you tried to solve. Explain in detail your approach, and specify any simplifications or assumptions you have taken. Also demonstrate the limitations of your approach. When doesn’t it work? Why? What steps would you have taken have you continued working on it? Make sure to add references to all related work you reviewed or used.
You are allowed to submit any supplementary material that you think it important to evaluate your work, however we do not guarantee that we will review all of that material, and you should not assume that. The report should be self-contained.
Submission: submit your report to stellar as a pdf file named <your_kerberos>.pdf. Submit any supplementary material as a single zip file named <your_kerberos>.zip. Add a README file describing the supplemental content.
Grading Policy
Final project occupies 40% of the course grade. The following is the weight for two parts:
- Research component of final project (30%)
- Abstract (3%)
- Introduction (3%)
- Related work (3%)
- Approach (and technical correctness) (6%)
- Experimental results (and technical correctness) (6%)
- Conclusion (2%)
- References (1%)
- Overall clarity of the report (3%)
- Reproducibility: can the work be reproduced from the information given in the report? (3%)
- Final presentation (10%)
- See the announcement email for grading rubrics.