Introduction to Computer Vision 2020/2021

Mathieu Aubry, Gül Varol, Karteek Alahari, Ivan Laptev

Course Information

Class time: Tuesdays 9:00 - 12:00

Room: E.Noether (ex UV)

News

Teaching Assistant: Robin Champenois, bonjour [at] robin-champenois [point] fr . Robin is your main contact for anything related to the programming assignments and final projects.

Pages from previous years: 2017-2018 ,2018-2019, 2019-2020

Evaluation
Assignments

There will be four programming assignments representing 60% of the grade. The supporting materials for the programming assignments projects will be in Python. You have to send your assignments to Robin by the deadline.

Final project

The final project will represent 40% of the grade. There can be two types of projects:

Presentation of a set of paper on a research topic (the list from last year is available here with more instructions, it will be slightly modified). Feel free to ask for papers on a topic that you are interested in or propose a paper (in this case, it has to be validated before the November 5th)
Working on a problem you define yourself. In this case, you should iterate on a 1-page proposal with Robin and myself BEFORE November 5th

Projects are in group of 2 or 3. You will have to present your project (~10 minutes x number of persons in the group + questions) and return a summary (2 pages max) of the essential points that should be readable (and useful) for the other students in the class.

Collaboration policy

You can discuss the assignments and final projects with other students in the class. Discussions are encouraged and are an essential component of the academic environment. However, each student has to work out their assignment alone (including any coding, experiments or derivations) and submit their own report. If you borrowed some code from somebody else, or wrote some code with somebody else, write it clearly in the report. Any uncredited reuse of material (text, code, results) will be considered as plagiarism and will result in zero points for the assignment. You all answer questions in very different ways, plagiarism is obvious when correcting assignments. If plagiarism is detected, the student will be reported to ENS.

Course schedule (subject to change):

Lecture	Date	Instructor	Topic and reading materials.	Slides	Syllabus
1	Sept. 22th	MA	Introduction, overview, Digital photography, projections and camera matrix refs : Forsyth and Ponce "Geometric camera model" chapter Szeliski chapter 2 "Image formation" Linear algebra for Vision (from Fei Fei Li): slides, pdf
2	Sept. 29st	MA	color, Human 3D vision/perception Edges (Canny) Linear and non-Linear Image filtering: Fourier and convolution, Bilateral Filter, Non-Local-Mean Ressources: relevant Book chapters on Fourier and linear image filtering (chapt. 2 and 3); Detailed presentation of Bilateral Filter ,
3	Oct. 6th	MA	Classical feature detectors (Harris Corners, blob detection) and descriptors, local features Instance recognition, RANSAC/Hough transform Projective geometry, Stereo vision printed introduction
4	Oct. 13th	MA	Camera calibration, multi-view reconstruction. Assignment 1 Due (Canny edges)
5	Oct. 20nd	MA	Segmentation (K-means, GMM, Mean shift) Optical Flow: optical flow equation, Lukas-Kanade, Horn and Schunk.
6	Nov. 3th	KA	Markov Random Fields: optimization methods (graph-cuts, belief propagation, TRW-S), applications to stereo and segmentation Assignment 2 Due (camera calibration)
7	Nov. 10th	KA	Markov Random Fields: optimization methods (graph-cuts, belief propagation, TRW-S), applications to stereo and segmentation
8	Nov. 17th	MA	Intro to Computer Graphics (intro to classical shape analysis) Introduction to category-level recognition / Introduction to CNNs Project choice finalized Assignment 3 Due (Mean Shift clustering)
9	Nov 24th	MA	Training CNNs Document on Stochastic Gradient Descent (Guillaume Obozinski)
10	Dec. 1rd	GV	visualizing/understanding CNN, Image classification, Transferring representations
11	Dec. 8th	GV	CNNs for object detection and semantic segmentation, more applications: 3D shape analysis/generation with NNs Assignment 4 Due (MNIST recognition with NN)
12	Dec 15th	IL	Low-level video analysis: tracking, human segmentation and pose
13	Jan. 5th	IL	High level video analysis, action recognition Project summary due
14	Jan. 12th	MA	Projects presentations
15	Jan. 19th	MA	Projects presentations

Relevant literature:

[1]	D.A. Forsyth and J. Ponce, "Computer Vision: A Modern Approach", Prentice-Hall, 2nd edition, 2011
[2]	J. Ponce, M. Hebert, C. Schmid and A. Zisserman "Toward Category-Level Object Recognition", Lecture Notes in Computer Science 4170, Springer-Verlag, 2007
[3]	O. Faugeras, Q.T. Luong, and T. Papadopoulo, "Geometry of Multiple Images", MIT Press, 2001.
[4]	R. Hartley and A. Zisserman, "Multiple View Geometry in Computer Vision", Cambridge University Press, 2004.
[5]	J. Koenderink, "Solid Shape", MIT Press, 1990
[6]	R. Szeliski, "Computer Vision: Algorithms and Applications", 2010. Online book.
[7]	Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT press. Online book

Good and relevant lectures by other people (many slides are taken from them)

James Hays https://www.cc.gatech.edu/~hays/compvision/

Svetlana Lazebnik http://slazebni.cs.illinois.edu/spring18/

Derek Hoeim https://courses.engr.illinois.edu/cs543/sp2017/

David Fouhey https://web.eecs.umich.edu/~fouhey/teaching/EECS442_W19/schedule.html