| Week | Suggested Reading | Lecture Notes |
|---|---|---|
| 1 | Buidling Rome in a Day project Their 2009 ICCV paper |
Introduction Overview |
| 2 | Szeliski 1.1, 1.2, 3.1, 3.2, 3.3 Wikipedia information about PNM, JPEG, and PNG image formats CVOnline on image transforms, especially Convolution (sec 7.3). |
Images and Filtering |
| 3 | Szeliski 4.1 A Combined Corner and Edge Detector, C. Harris and M. Stephens, Proc. 4th Alvey Vision Conf., 147-151, 1988 Distinctive Image Features from Scale-Invariant Keypoints, D. Lowe, IJCV 60(2): 91-110, 2004 A Comparison of Affine Region Detectors, K. Mikolajczyk et al., IJCV 65(1/2): 43-73, 2005 |
Feature Detection |
| 4 | Szeliski 4.1 Distinctive Image Features from Scale-Invariant Keypoints, D. Lowe, IJCV 60(2): 91-110, 2004 A Performance Evaluation of Local Descriptors, K. Mikolajczyk and C. Schmid, IEEE PAMI 27(10): 1615-1630, 2005 |
Feature Description |
| 5 | Szeliski 2.1 The Eigen matrix library |
Cameras and Transforms |
| 6 | Szeliski 6.1 Hartley and Zisserman, Chapter 4 A short introduction to the Levenberg-Marquardt algorithm |
Estimating Transforms |
| 7 | Szeliski 9 Image Alignment and Stitching: A Tutorial, R. Szeliski, Microsoft Technical Report MSR-TR-2004-92, 2004 Global Seamline Networks for Orthomosaic Generation via Local Search, S. Mills and P. McLeod, JPRS 27: 101-111, 2013 Also, seminar presenting this paper |
Image Mosaicing |
| 8 | Szeliski 7.2, 11.1-3 Hartley and Zisserman Chapter, 9 - available online |
Stereo Geometry |
| 9 | Szeliski 7.1, 6.2 Hartley and Zisserman Chapter, 9 - as for last lecture, particularly 9.6 Random sample consensu M. A. Fischler and R. C. Bolles, Communications of the ACM 24(6): 381-395, 1981 |
3D Motion and Structure |
| 10 | Szeliski 7.4 Bundle Adjustment - A Modern Synthesis B. Triggs, P. McLauchlan, R. Hartley, and A. Fitzgibbon, Vision Algorithms: Theory and Practice, 298-397, 2000 SBA: A Software Package for Generic Sparse Bundle Adjustment M.I. A. Lourakis and A.A. Argyros, ACM Trans. Math. Software 36(1): 1-30, 2009 There is also an implementation available for this. Alternatively there is a recent package called Ceres which does a range of non-linear least squares problems. |
Bundle Adjustment |
| 11 |
Microsoft's Kinect for Windows page. Kinect Fusion: Real-time 3D Reconstruction and Interaction Using a Moving Depth Camera S. Izadi et al., Proc. ACM Symposium on User Interface Software and Technology, 559-568, 2011 Shake n Sense: Reducing Interference for Overlapping Structured Light Depth Cameras D.A. Butler et al., Proc ACM Conf. Human Factors in Computing Systems, 1933-1936, 2012 Source code for the demo in lectures - note that this requires OpenCV, the Kinect for Windows SDK, a Kinect for Windows device, but gives an idea of the use of the SDK. 3D Photography on Your Desk J.-Y. Bouguet and P. Perona, Int. Conf. Computer Vision, 43-50, 1998. Not Kinect-related but a simpler example of structured light. |
RGB-D Sensing |
| 12 |
The Middlebury Stereo Vision Page Accurate, Dense, and Robust Multi-View Stereopsis, Y. Furukawa and J. Ponce, IEEE Trans. Pattern Analysis and Machine Intelligence, 32(8), 1362-1376, 2010 Poisson Surface Reconstruction. M. Kazhdan, M. Boltiho, and H. Hoppe, Symposium on Geometry Processing, 61-70, 2006 A sample examination paper is available to help you prepare for the real thing. |
Surface Reconstruction Revision + the Exam |