|
|
COS 429 - Computer Vision |
Fall 2019 |
| Course home | Outline and Lecture Notes | Assignments |
The lecture slides are provided for reference but are subject to change. We will update the pdf posted here with any major changes, but otherwise expect minor changes between what's posted here and the lecture.
| Date | Lecture (click for slides) | Readings | Assignments |
|---|---|---|---|
| Thu, Sep 12 | Introduction to computer vision | Ch. 1.1 | |
| Tue, Sep 17 | Image formation and capture (slide 59 updated 10/16) | Ch. 2.2.3, 2.3 | |
| Thu, Sep 19 | Convolution and filtering | Ch. 3.2, 3.4, 4.2 | Assignment 0 due |
| Tue, Sep 24 | Feature detectors and descriptors | Ch. 4, Trucco & Verri: Ch. 4.1 – 4.3, SIFT paper | |
| Thu, Sep 25 | Fitting, Hough transforms, RANSAC | Ch. 4.3.2, 6.1.4, COS 324 notes on linear regression | |
| Tue, Oct 1 | Image alignment and stitching | Ch. 6.1.1 – 6.1.4, 9, Multires blending paper | |
| Thu, Oct 3 | Intro to recognition and machine learning (slides updated 10/5 based on lecture content) | COS324 notes on least squares, regularization, cross-validation | Assignment 1 due |
| Tue, Oct 8 | Face and pedestrian detection(slides updated 10/8 post-lecture) | Ch. 14.1, Viola & Jones paper, Dalal & Triggs paper, COS324 notes on SVM optimization, GenderShades paper | |
| Thu, Oct 10 | Object classification | Ch. 14.4.1, 14.5, 14.6, Caltech 101 paper, SPM paper, COS324 notes on kmeans | |
| Tue, Oct 15 | Object detection | Ch. 14.4.2, Deformable Parts Model (DPM) paper, PASCAL VOC paper, Diagnosing Errors paper, ImageNet Challenge paper | |
| Thu, Oct 17 | Segmentation (slide 32 updated 10/18) | Ch. 5.2 – 5.4;
Martin et al. segmentation paper;
Shi and Malik normalized cuts paper |
Assignment 2 due |
| Tue, Oct 22 | Texture | Ch. 10.5;
Efros & Leung paper; Efros & Freeman paper; Image Quilting |
|
| Thu, Oct 24 | Midterm | ||
| Tue, Oct 31 | No class - fall break | ||
| Thu, Nov 2 | No class - fall break | ||
| Tue, Nov 5 | Motion, optical flow, tracking (slide 19 updated following lecture) | Ch. 8.4; Lucas-Kanade paper;
Ch. 8.1 – 8.3 (optional) |
|
| Thu, Nov 7 | 3D vision: stereo, camera geometry and calibration | Ch. 11.1 – 11.4,11.6; Ch. 12 (optional); Stanford CS 231A course notes on epipolar geometry | |
| Tue, Nov 12 | Introduction to Deep Learning | ||
| Thu, Nov 14 | Deep learning II: Backpropagation | ||
| Tue, Nov 19 | Deep learning III: CNNs and ImageNet | CNN Beginner's Guide | |
| Thu, Nov 21 | Deep Learning IV: Training | Assignment 3 due | |
| Tue, Nov 26 | Deep Learning V: Training and architectures | ||
| Thu, Nov 28 | No class - Thanksgiving | ||
| Tue, Dec 3 | Deep Learning VI: Advanced recognition topics | ||
| Thu, Dec 5 | William Pierson Field Lecture: Dr. Juan Carlos Niebles (Stanford) on "Human Event Understanding: From Actions to Tasks" | Assignment 4 due | |
| Tue, Dec 10 | Advanced topics: Fairness in computer vision | ||
| Thu, Dec 12 | Guest Lecture: Dr. Andras Ferencz (Mobileye) on "Computer vision for autonomous driving" | ||
| Fri, Dec 13 | Project milestone due | ||
| TBA | Final project presentations | ||
| Tue, Jan 14 | Project report due | ||