**CS 5840: Computer Vision**

Fall 2017

Meets MW 4:45-6:00 PM, OSB B216

**Syllabus:** PDF**Blackboard:** http://bb.uccs.edu

The gradebook and homework turn-in pages will be maintained in Blackboard.

**Computer access**

UCCS EAS maintains physical and virtual machines for use by students. Please see the UCCS EAS IT Helpdesk for more information.

We have several machines available with Nvidia GPUs suitable for deep learning tasks. Currently online are cs-gpuN.uccs.edu where N = 1,2,3, or 4 (GTX 1050 Ti).

**Textbooks**

R. Szeliski. Computer vision: algorithms and applications. Springer, 2010.

The textbook is available online for free as a PDF download at Richard Szeliski’s website, and is also available for purchase at the UCCS Bookstore.

S. Palmer. Vision Science: Photons to Phenomenology. MIT Press, 1999.

This is available electronically through the library.

I. Goodfellow, Y. Bengio, and A. Courville. Deep Learning. MIT Press, 2016.

Available online.

**Interesting Links**

- Deep Art
- You Look Familiar: Unearthing the Face Within: Doris Tsao at TEDxCaltech
- How We Save Face--Researchers Crack the Brain's Facial-Recognition Code
- Image Resizing by Seam Carving
- New Theory Cracks Open the Black Box of Deep Learning

**Introductory Material (from Deep Learning book)**

**Class Schedule**

*Monday, August 21:*Introduction to Computer Vision*Wednesday, August 23:*Human Visual System; Python Tutorial- Reading: Palmer sections 1.2.3-1.3.3
- Slides
*Monday, August 28:*Image Formation- Reading: Szeliski section 2.1
- Slides
*Wednesday, August 30:*Geometric Transformations- Reading: Szeliski section 2.1
- Slides
*Wednesday, September 6:*Filtering- Reading: Szeliski section 3.1-3.2
- Slides
*Monday, September 11:*Thinking in Frequencies- Reading: Szeliski section 3.3-3.4
- Slides
*Wednesday, September 13:*Image Pyramids and Edges- Reading: Szeliski section 3.5, 4.2
- Paper: P. J. Burt and E. H. Adelson. The Laplacian Pyramid as a Compact Image Code. IEEE Transactions on Communications, 1983.
- Slides
*Monday, September 18:*Lines, Corners and Blobs- Reading: Szeliski section 4.1, 4.3
- Slides
*Wednesday, September 20:*Feature Matching and Tracking- Reading: Szeliski section 4.1
- Paper: D. Lowe. Distinctive Image Features from Scale-Invariant Keypoints. IJCV 2004
- Slides
*Monday, September 25:*Eigenfaces and Active Appearance Models- Reading: Szeliski section 14.2
- Slides
*Wednesday, September 27:*Introduction to machine learning, nearest neighbor and linear classification- Reading: Machine Learning Basics from Deep Learning
- Slides: Intro to Machine Learning
- Slides: Neural Networks: Theory
*Monday, October 2:*Machine Learning: Optimization Methods- Readings:
- Reading: Machine Learning Basics from Deep Learning
- Lecture notes on optimization and backpropagation from Stanford CS231N
- Slides: Neural Networks (Optimization)
*Wednesday, October 4:*Neural Networks- Readings:
- Chapter 6: Deep Feedforward Networks from Deep Learning
- Chapter 7: Regularization for Deep Learning from Deep Learning
- Chapter 8: Optimization for Training Deep Models from Deep Learning
- Supplementary: Lecture notes (1, 2, 3, 4) on neural networks from Stanford CS231N
- Slides: Theory and Optimization
*Monday, October 16:*Convolutional Neural Networks- Readings:
- Chapter 9: Convolutional Networks from Deep Learning
- Supplementary: Lecture notes (1, 2, 3) on convolutional neural networks from Stanford CS231N
- Slides: Convolutional Neural Networks
*Wednesday, October 18:*Optical Flow and Epipolar Geometry- Reading: Szeliski section 7.2
- Slides
*Monday, October 23:*Structure-from Motion and Bundle Adjustment- Reading: Szeliski section 7.4
- Slides