1 | General Information | |||
---|---|---|---|---|
2 | Course Number: | CS 6476-O01 | ||
3 | Course Name: | Computer Vision | ||
4 | Program: | Georgia Tech's Online MS in Computer Science | Link to GT OMS-CS Website | |
5 | ||||
6 | Team/People | |||
7 | Instructor | Irfan Essa | Irfan Essa's Home Page | |
8 | Contact via Piazza. Email for Private and Urgent Issues ONLY | Piazza Site | ||
9 | Video Lectures by | Aaron Bobick | Aaron Bobick's Home Page | |
10 | ||||
11 | ||||
12 | TAs | Robert Capps | Contact via Piazza. Email for Private and Urgent Issues ONLY, Use Post to Instructors OPTION in Piazza | |
13 | Paritosh Gote | |||
14 | Will Hancock | |||
15 | Aditya Kadur | |||
16 | Sarah Moore | |||
17 | ||||
18 | Instructional Designers | Arpan Chakraborty | ||
19 | Video Production | Megan Smith | ||
20 | ||||
21 | Assignments & Grading | |||
22 | A. | Assignments / Homeworks Type 1: (There will be 1 assignments of this type) | 4.0% | 4% Each |
23 | B. | Assignments / Homeworks Type 2: (There will be 7 assignments of this type) | 77.0% | 11% Each |
24 | C. | Exam: [Scheduled two weeks before end of term, cumulative, and online] | 15.0% | |
25 | F. | Participation on Piazza (Measured by Activity on Piazza) | 4.0% | Entire Term |
26 | ||||
27 | Total | 100.0% | ||
28 | ||||
29 | ||||
30 | Policies | |||
31 | ||||
32 | Communications | |||
33 | A. | WITH the Professor and TAs should be exclusively through Piazza. No emails! Professor and TAs will do their best to respond to questions within 2 days of posted question. | ||
34 | B. | Piazza will serve as the primary and ONLY source of communications and sharing announcements with the students. | Piazza Site for this class | |
35 | C. | All communications should be professional and courteous. TA/Graders and Students are all required to maintian high standards of interaction on Piazza | Piazza Site for this class | |
36 | ||||
37 | Assignments | |||
38 | A. | T-square will be used for all assignment submissions and grading. Dates and Deadlines in T-square are the final authority | T-square Site for this class | |
39 | B. | Homeworks Assignments will be graded on a list of criteria (specified on the assignment) such as quality of work, completeness, insight into technical issues, insight into other relevant issues, etc. | ||
40 | C. | Each assignment will be graded and returned USUALLY within two weeks of submission. If there is delay for some reason, it will be announced. | ||
41 | D. | Late Assignments: Everything is DUE when specified. NO extensions. Each day of LATE submission will result in a 10 point penalty, for up to 2 days, after which you will receive a 0 ! | ||
42 | E. | See collaboration policy below for more details on how to collaborate | ||
43 | F. | Instruction provided with the assignment, MUST be explicitly followed, especially any and all directions like how to submit and the file naming conventions specified | ||
44 | G. | Regrade requests can be made using the Google Form (on the right). Please provide clear details as to why you are requesting a regrade. All regrade requests must be made within TWO (2) weeks of the grade release. For grades released in the last two weeks of the term, the regrade request must be made by the last day of the final exams week. | Regrade Request Form | |
45 | F. | All DUE dates will be on the T-square site, and the timezone will be ATLANTA (US-EST) time. Please plan accordingly | ||
46 | ||||
47 | Websites | Following are the websites we will OFFICIALLY use for this class. | ||
48 | A. | T-square: For Assignment Submission, Grading, and Final Exam. | T-square | |
49 | B. | Piazza: For Official Announcements, Forums for discussion. | Piazza | |
50 | C. | WordPress Site: (This site) for syllabus/schedule and general information. | Worpdress | |
51 | D. | Udacity for videos of lectures. | udacity.com | |
52 | E. | No information will be shared via any other site (G+, FB, etc.). Students are welcome to create their own social media sites, but none of the instructors are required to be on those sites and will not participate there regularly. | ||
53 | G. | As we have a 8 assignments, there will be overlap on assignments. We expect students to manage their schedule to meet the deadlines for each of the assignments | ||
54 | H. | Students are welcome to work and submit assignments before their due date. The lectures will all be available from week 1. TAs will try to answer questions related to the assignments as much as they can, but most conversations maybe most active as per the Schedule planned for the class | ||
55 | ||||
56 | Grading | Grading Scale (for each assignment/unit and for the entire class). | T-square | |
57 | A | Above 90% | ||
58 | B | 80%-89.99% | ||
59 | C | 70%-79.99% | ||
60 | D | 60%-69.99% | ||
61 | F | Below 60% | ||
62 | Note: Any work that meets all the requirements will be given a 90%. For scores above 90%, work has to above and beyond meeting the basic requirements of the assigned work. | |||
63 | ||||
64 | Honor Code | All assigned work is expected to be individual, except where explicitly written otherwise. You are encouraged to discuss the assignments with your classmates; however, what you hand in should be your own work. If any work product was produced based on discussions with someone else (in the class OR outside), please specify clearly in the final turn-in. | GT Honor Code | |
65 | ||||
66 | Collaboration Policy | As stated above with the Honor Code, but worth making explicit here. Collaboration between students on work assigned in class is fine. You are encouraged to discuss your work with each other. But each individual students MUST submit their own work, done solely by themselves. In some cases, you may have had a fellow student or a non-student friend, help you with an assignment or work (say to take a picture!). You are REQUIRED to acknowledge any help you may have received in completing the work assigned, even as small as holding the light, or suggesting a possible path to a solution. Please be explicit and provide details. We will be checking for code plagarism in our assessment, so please NO copying code from the Web/Internet |
1 | Week OF | Lectures | Assignments | Readings | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2 | # | Begin | Module | Title | Lesson | Topic | Duration (min) | PS# | Title | Out (See T-Sq) | Due (See T-Sq) | Slides | Text | Other | |
3 | DUE DATES provided here for just for planning purposes. ACTUAL due dates are in T-Square. Time Zone for this class is Atlanta Time (EST). | ||||||||||||||
4 | 1 | 17-Aug | 1A | Introduction | 1A-L1 | Introduction | 42 | PS1 | Images as Functions | 17-Aug | 31-Aug | Install Python, OpenCV | |||
5 | 2A | Linear image processing | 2A-L1 | Images as functions | 44 | ||||||||||
6 | 2A-L2 | Filtering | 24 | ||||||||||||
7 | 2A-L3 | Linearity and convolution | 34 | ||||||||||||
8 | 2A-L4 | Filters as templates | 13 | ||||||||||||
9 | 2A-L5 | Edge detection: Gradients | 27 | ||||||||||||
10 | 2A-L6 | Edge detection: 2D operators | 19 | ||||||||||||
11 | 2 | 24-Aug | 2B | Model fitting | 2B-L1 | Hough transform: Lines | 36 | PS2 | Edges and Lines | 24-Aug | 8-Sep | ||||
12 | 2B-L2 | Hough transform: Circles | 13 | ||||||||||||
13 | 2B-L3 | Generalized Hough transform | 16 | ||||||||||||
14 | 2C | Frequency domain analysis | 2C-L1 | Fourier transform | 36 | ||||||||||
15 | 2C-L2 | Convolution in frequency domain | 22 | ||||||||||||
16 | 2C-L3 | Aliasing | 35 | ||||||||||||
17 | 3 | 31-Aug | 3A | Camera models | 3A-L1 | Cameras and images | 33 | PS3 | Window-based Stereo Matching | 31-Aug | 14-Sep | ||||
18 | 3A-L2 | Perspective imaging | 26 | ||||||||||||
19 | 3B | Stereo geometry | 3B-L1 | Stereo geometry | 26 | ||||||||||
20 | 3B-L2 | Epipolar geometry | 11 | ||||||||||||
21 | 3B-L3 | Stereo correspondence | 29 | ||||||||||||
22 | 4 | 7-Sep | 3C | Camera calibration | 3C-L1 | Extrinsic camera calibration | 24 | PS4 | Geometry | 8-Sep | 21-Sep | ||||
23 | 3C-L2 | Instrinsic camera calibration | 16 | ||||||||||||
24 | 3C-L3 | Calibrating cameras | 31 | ||||||||||||
25 | 5 | 14-Sep | 3D | Multiple views | 3D-L1 | Image to image projections | 10 | ||||||||
26 | 3D-L2 | Homographies and mosaics | 33 | ||||||||||||
27 | 3D-L3 | Projective geometry | 14 | ||||||||||||
28 | 3D-L4 | Essential matrix | 22 | ||||||||||||
29 | 3D-L5 | Fundamental matrix | 37 | ||||||||||||
30 | 6 | 21-Sep | 4A | Feature detection | 4A-L1 | Introduction to "features" | 13 | PS5 | Harris, SIFT, RANSAC | 21-Sep | 5-Oct | ||||
31 | 4A-L2 | Finding corners | 39 | ||||||||||||
32 | 4A-L3 | Scale invariance | 23 | ||||||||||||
33 | 4B | Feature descriptors | 4B-L1 | SIFT descriptor | 27 | ||||||||||
34 | 4B-L2 | Matching feature points (a little) | 16 | ||||||||||||
35 | 4C | Model fitting | 4C-L1 | Robust error functions | 31 | ||||||||||
36 | 4C-L2 | RANSAC | 33 | ||||||||||||
37 | 7 | 28-Sep | 5A | Photometry | 5A-L1 | Photometry | 35 | ||||||||
38 | 5B | Lightness | 5B-L1 | Lightness | 26 | ||||||||||
39 | 5C | Shape from shading | 5C-L1 | Shape from shading | 34 | ||||||||||
40 | 8 | 5-Oct | 6A | Overview | 6A-L1 | Introduction to motion | 16 | PS6 | Optic Flow | 5-Oct | 20-Oct | ||||
41 | 6B | Optical flow | 6B-L1 | Dense flow: Brightness constraint | 24 | ||||||||||
42 | 6B-L2 | Dense flow: Lucas and Kanade | 17 | ||||||||||||
43 | 6B-L3 | Hierarchical LK | 33 | ||||||||||||
44 | 6B-L4 | Motion models | 24 | ||||||||||||
45 | 9 | 12-Oct | 7A | Introduction to tracking | 7A-L1 | Introduction to tracking | 14 | ||||||||
46 | 7B | Parametric models | 7B-L1 | Tracking as inference | 21 | ||||||||||
47 | 7B-L2 | The Kalman filter | 36 | ||||||||||||
48 | 10 | 19-Oct | 7C | Non-parametric models | 7C-L1 | Bayes filters | 23 | PS7 | Particle Tracking | 20-Oct | 9-Nov | ||||
49 | 7C-L2 | Particle filters | 17 | ||||||||||||
50 | 7C-L3 | Particle filters for localization | 24 | ||||||||||||
51 | 7C-L4 | Particle filters for real | 15 | ||||||||||||
52 | 11 | 26-Oct | 7D | Tracking considerations | 7D-L1 | Tracking considerations | 27 | ||||||||
53 | 8A | Introduction to recognition | 8A-L1 | Introduction to recognition | 21 | ||||||||||
54 | 8B | Classification: Generative models | 8B-L1 | Classification: Generative models | 28 | ||||||||||
55 | 8B-L2 | Principle Component Analysis | 48 | ||||||||||||
56 | 8B-L3 | Appearance-based tracking | 26 | ||||||||||||
57 | 12 | 2-Nov | 8C | Classification: Discriminative models | 8C-L1 | Classification: Discriminative models | 27 | ||||||||
58 | 8C-L2 | Boosting and face detection | 27 | ||||||||||||
59 | 8C-L3 | Support Vector Machines | 51 | ||||||||||||
60 | 8C-L4 | Bag of visual words | 14 | ||||||||||||
61 | 13 | 9-Nov | 8D | Action recognition | 8D-L1 | Introduction to video analysis | 24 | PS8 | Motion History Images | 9-Nov | 23-Nov | ||||
62 | 8D-L2 | Activity recognition | 32 | ||||||||||||
63 | 8D-L3 | Hidden Markov Models | 46 | ||||||||||||
64 | 14 | 16-Nov | 9A | Color spaces and segmentation | 9A-L1 | Color spaces | 36 | ||||||||
65 | 9A-L2 | Segmentation | 18 | ||||||||||||
66 | 9A-L3 | Mean shift segmentation | 18 | ||||||||||||
67 | 9A-L4 | Segmentation by graph partitioning | 13 | ||||||||||||
68 | 15 | 23-Nov | 9B | Binary morphology | 9B-L1 | Binary morphology | 37 | ||||||||
69 | 9C | 3D perception | 9C-L1 | 3D perception | 34 | ||||||||||
70 | 10A | The retina | 10A-L1 | The retina | 38 | ||||||||||
71 | 16 | 30-Nov | 10B | Vision in the brain | 10B-L1 | Vision in the brain | 27 | ||||||||
72 | 17 | 1-Dec | Final Exam (T-Square) | 26-Nov | 9-Dec | Guide |