Title image Fall 2019

Syllabus for Fall 2019

Topics and Reading Assignments


We will be using primarily technical papers as the basis for the course. Computer vision papers will be mostly from CVPR, ICCV, ECCV, TPAMI, and IJCV.

There are few relevant textbooks, as many of the topics we will be exploring are current research. The Deep Learning textbook by Goodfellow, Bengio, and Courville is a helpful general resource. Likewise, the Computer Vision textbook by Szeliski is also useful for symbolic computer vision algorithms.


Weekly critiques and presentations 30%
Participation and attendance 10%
3 Primary Projects 40%
Final project 20%

This course is a combination of a weekly seminar-style discussion and four significant projects. The four projects are given on the assigment page.

In a typical week we will go over 5-6 technical papers on a selected topic. The papers will span from the algorithm's initial development to recent modifications and updates. Everyone will be expected to have at least skimmed all of the papers.

Each paper will have two people selected to present it to the group; You can expect to present every other week. The presenters are responsible for writing a short summary and critique of the paper--generally one page--that summarizes the algorithm and identifies points for discussion. The critiques need to be posted at least 12 hours prior to meeting time. The critique will help guide the discussions during the meeting time.

Late Policy: All paper critiques must be available to the rest of the class at least 12 hours before the class meeting where the paper will be discussed. Late critiques will not receive full credit.

Students will have three weeks for each of the first three projects. Each student will present their project in the class following the due date. The presentation is worth 20% of the project grade. Reduced credit will be given for late projects.

Starting a project in the last few days before it is due will guarantee a poor project and a poor grade. The projects are three week projects because you should plan on them taking between 20-25 hours of study, design, and programming. Spread across three weeks, that is a reasonable 6-8 hours per week.

Attendance: this course involves presenting, discussing, and engaging the material with the other people in the class. If you miss a class, you cannot make up those parts of the course. Missing class will result in a reduced participation grade.


Use the college wiki to post your critiques and to hand in your project reports. Your project reports should contain lots of pictures and an overall description of your design, algorithm, and implementation.

For each critique or assignment, give the page the label specified in the assignment so the professor and other students can find them. For the critiques, please use the label format cs465f19.

Weekly Topics and Readings

  • Illuminant Estimation
  • Camera Mapping
  • Color Correction
  • Super-resolution
  • Blur Estimation
  • Noise Removal
  • Shadow Removal
  • Intrinsic Imaging
  • Speech Recognition
  • HMMs
  • Project 1 presentations
  • Speech Recognition II
  • Recurrent nets and LSTMs
  • Machine Translation
  • Sign Language Recognition
  • Video Analysis
  • Video Prediction
  • Project 2 presentations
  • Human Pose Estimation
  • Activity Recognition I
  • Activity Recognition II
  • Video Summarization
  • Project 3 presentations
  • Depth
  • Detecting 3D Objects

Collaboration and Academic Honesty

Computer science, both academically and professionally, is a collaborative discipline. In any collaboration, however, all parties are expected to make their own contributions and to generously credit the contributions of others. In our class, therefore, collaboration on homework and programming assignments is encouraged, but you as an individual are responsible for understanding all the material in the assignment and doing your own work. Always strive to do your best, give generous credit to others, start early, and seek help early from both your professors and classmates.

In addition to the ethical implications of dishonesty, you undermine your ability to learn when you cheat. Honesty, integrity, and personal responsibility are cornerstones of a Colby education and provide the foundation for scholarly inquiry, intellectual discourse, and an open and welcoming campus community. These values are articulated in the Colby Affirmation and are central to this course. Students are expected to demonstrate academic honesty in all aspects of this course.

The following rules are intended to help you get the most out of your education and to clarify the line between honest and dishonest work. We reserve the right to ask you to verbally explain the reasoning behind any answer or code that you turn in and to modify your project grade based on your answers. It is vitally important that you turn in work that is your own. We do use automated plagiarism detection software, so please be sure to abide by these, rather minimal, rules. Reports of academic dishonesty are handled by an academic review board and a finding of academic dishonesty may result in significant sanctions. For more details on Colby's Academic Integrity policies and procedures, see colby.edu/academicintegrity.

Sexual Misconduct/Title IX Statement

Colby College prohibits and will not tolerate sexual misconduct or gender-based discrimination of any kind. Colby is legally obligated to investigate sexual misconduct (including, but not limited to sexual assault and sexual harassment).

If you wish to speak confidentially about an incident of sexual misconduct, please contact Colby Counseling Services (207-859-4490) or the Director of the Gender and Sexual Diversity Program, Emily Schusterbauer (207-859-4093).

Students should be aware that faculty members are considered responsible employees; as such, if you disclose an incident of sexual misconduct to a faculty member, they have an obligation to report it to Colby's Title IX Coordinator. "Disclosure" may include communication in-person, via email/phone/text, or through class assignments.

To learn more about sexual misconduct or report an incident, visit http://www.colby.edu/sexualviolence/.

The Colby Affirmation

Colby College is a community dedicated to learning and committed to the growth and well-being of all its members.

As a community devoted to intellectual growth, we value academic integrity. We agree to take ownership of our academic work, to submit only work that is our own, to fully acknowledge the research and ideas of others in our work, and to abide by the instructions and regulations governing academic work established by the faculty.

As a community built on respect for ourselves, each other, and our physical environment, we recognize the diversity of people who have gathered here and that genuine inclusivity requires active, honest, and compassionate engagement with one another. We agree to respect each other, to honor community expectations, and to comply with College policies.

As a member of this community, I pledge to hold myself and others accountable to these values.

  • We respect each other and ourselves.
  • We respect our physical spaces on campus.
  • We respect our academics and complete work honestly.