Requirements:

Students are expected to do the assigned reading, participate in class discussions, write one paper review each week, present a paper/topic in a class and complete a final project.  A paper presentation involves doing background research on a topic. 


Note that presentations are **one week before** the slot your presentation is scheduled.  This deadline is a hard deadline.  This means you will need to read the papers, prepare experiments (optional), slides, etc. one week before the date you are signed up for.  The idea is to meet and discuss ahead of time, so that we can iterate as needed the week leading up to your presentation. 

 

Paper reviews

 The quality of our discussions will rely on how prepared everyone is when they come to class.  It is important to do the reading in order to actively participate.  Students are required to submit one paper reviews per week for the assigned papers.  (We'll usually read 2 papers each week; just choose 1 of those assigned to review.)

Each review should address the following (in any order):

o    Give a summary of the paper in your own words (very brief, 2-3 sentences)

o    What is the main contribution of the paper?

o    What are the primary strengths and weaknesses of the paper?

o    How convincing are the experiments?  If something specific is lacking, what should have been tested?

o    Describe one specific way in which the work could be extended (bonus).

o    Additional comments, including unclear points, connections you see between the papers

.

See examples of well-written reviews.

Reviews are due by 9 PM on the night before class (Sunday).  Email reviews to me, pasting the text directly into your mail (no attachments, please).  Include [4738] in the subject header.

In weeks that you are presenting, you can skip writing the reviews.

 

Paper presentations

 

Each team (of 2 students) will give a presentation in class covering 2 papers on a topic selected from the course syllabus list.  This presentation should overview the papers and explain technical details, and synthesize any underlying commonalities or highlight interesting distinctions.
 

The talk should be well-organized and polished, sticking to about 40 minutes (20 min.each student).  Please run through it beforehand and check the time (a good rule of thumb: generally 20 minutes ~ target 20 slides total).

Include these components in the presentation:

o    Clear statement of the problem

o    Why the problem is interesting, important, difficult

o    Key technical ideas, how they work, main contributions, strengths and weaknesses

o    Evaluation, summary of key experiments and data

o    Open issues raised in the papers, likely extensions

 

Try to use applications to motivate the work when possible, and look for visual elements (images, videos) to put in the presentation.  Check out the webpages linked on the class webpage, and also look at authors webpages for supplementary materials.  It's ok to grab a few slides from conference talks etc. when available, but be sure to clearly cite the source on *each* slide that is not your own. 

 Here is an example of a good presentation: example 2.

  

Projects

 

A project could be built around any of the following, and should be done with a partner. Experimental evaluation should be done on a benchmark data set ( provided in the course page)

 

         Object recognition using bag-of-words representation and discriminative training.

         Face detection

         Pedestrian detection

         Face recognition

         Saliency

         Action recognition

         Image retrieval

         Any other recognition application we discussed in class.

You can use papers provided as an additional reading to choose your project.
Initial project proposals will be due before the middle of the term. 

 

Grading:

Grades will be determined as follows:

         30% participation (includes attendance, in-class discussions, paper reviews)

         40% presentations (in-class presentation)

         30% final project (implementaion, presentation, final paper)