Week 1. Introduction
Week 2. Guest Lecture
Week 3. Image Filtering
Week 4. Features
Week 5. Model Fitting
Week 6-7. Camera Model
Week 7. Homography
Week 8. Stereo
- Fun with homography: PDF
- David Kriegman's notes on homography estimation: PDF
Week 8 RGBD image processing
Week 9 Segmentation
Week 10 Bag of Words and Image Search
Week 11 Motion
- Stereo. Methods to find
correspondence between two stereo images:
local search, dynamic programming, max-flow.
- Optical Flow. Lucas-Kanade Algorithm.
Least square method. Opencv function.
- ICM (Iterative conditional mode): a local search method.
- DP on trees. Decomposition method.
Week 11: Tracking
Week 12: Action Recognition
Week 13 Human Pose
- A. Efros, A. Berg, G. Mori, J. Maliak
"Recognizing Actions at a Distance", ICCV '03.
- E. Shechtman and M. Irani, Space-time behavior based correlation - OR -
How to tell if two underlying motion fields are similar without computing them? IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 29(11): 2045-2056, November 2007.
- L. Gorelick, M. Blank, E. Shechtman, M. Irani, and R. Basri, Actions as Space-Time Shapes. In IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 29(12): 2247-2253, December 2007.
- Action recongnition using space-time interest point.
- Action detection using trajectories.
Week 14-15 Classification Methods in Computer Vision
- P. Felzenszwalb, D. Huttenlocher,
"Efficient Matching of Pictorial Structures", CVPR 2000.
- D. Ramanan, D. A. Forsyth, A. Zisserman. "Strike a Pose: Tracking People by Finding Stylized Poses", CVPR 2005.
- Hao Jiang,
Human Pose Estimation Using Consistent Max-Covering
IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33, no. 9, 2011.
Finding poses in sports videos.
- C. J. Taylor, "Reconstruction of Articulated Objects from Point Correspondences in a Single Uncalibrated Image",
Computer Vision and Image Understanding, Vol: 80, No: 10, Pgs: 349-363, October 2000
Week 16 Final Project Mid-term Presentation
- Nearest neighbor, K-nearest neighbor, Approximate nearest neighbor.
- How to build a tree classifier: entropy and information gain.
- Face detection, Pedestrian detection.
- Recognizing deformable objects.