<html><head><style type='text/css'>p { margin: 0; }</style></head><body><div style='font-family: arial,helvetica,sans-serif; font-size: 10pt; color: #000000'>Sema Berkiten will present her research seminar/general exam on Monday May 14 at 10AM in <br>Room 402. The members of her committee are: Szymon Rusinkiewicz (advisor), Adam <br>Finkelstein, and Kai Li. Everyone is invited to attend her talk and those faculty wishing to <br>remain for the oral exam following are welcome to do so. Her abstract and reading list <br>follow below.<br><br><hr id="zwchr"><div style="color:#000;font-weight:normal;font-style:normal;text-decoration:none;font-family:Helvetica,Arial,sans-serif;font-size:12pt;"><b></b><div class="gmail_quote"><p></p><p><b><br>Abstract:</b></p><p>Image alignment is one of the very first steps for most computer vision algorithms. Image fusion, image mosaicing, image panorama, object recognition and detection, photometric stereo enhanced rendering are some of the examples in which image alignment is a crucial step for a promising result. In this work, we focused on a specific problem which is alignment of high-resolution images taken from the same viewpoint under different light directions. Although images are taken from the same viewpoint, there might be some misalignment due to perturbations to the camera and the effect of optical image stabilization. For this specific alignment problem, we made a broad literature survey and try different tools and algorithms to compare different approaches and find out the best approach to solve this problem. Based on our experiments, we saw that the best approaches are the feature-based ones. We found SIFT and SURF reliable for most cases. For feature-based approaches, one of the main problems is elimination of outliers, and we solved this problem with using RANSAC framework. Also datasets that we focus on have many images, between 30-60, of the same object, and in order to take advantage of having many images, we constructed a graph on images so that the most similar images are connected. Minimum spanning tree is one way of constructing such a data structure. As a future work we are planning to explore how to distribute the accumulated alignment error because of the tree structure. <br></p><p><br></p>
<p></p><p><b>Reading List:</b></p>
<p>1. Chunming Tang; Yan Dong; Xiaohong Su; , "Automatic Registration Based on Improved SIFT for Medical Microscopic Sequence Images," Intelligent Information Technology Application, 2008. IITA '08. Second International Symposium on , vol.1, no., pp.580-583, 20-22 Dec. 2008</p>
<p>2. Matthew Brown and David G. Lowe. 2007. Automatic Panoramic Image Stitching using Invariant Features. Int. J. Comput. Vision 74, 1 (August 2007), 59-73. DOI=10.1007/s11263-006-0002-3 <span><a href="http://dx.doi.org/10.1007/s11263-006-0002-3" target="_blank">http://dx.doi.org/10.1007/s11263-006-0002-3</a></span></p>
<p>3. David G. Lowe. 2004. Distinctive Image Features from Scale-Invariant Keypoints. Int. J. Comput. Vision 60, 2 (November 2004), 91-110. DOI=10.1023/B:VISI.0000029664.99615.94 <span><a href="http://dx.doi.org/10.1023/B:VISI.0000029664.99615.94" target="_blank">http://dx.doi.org/10.1023/B:VISI.0000029664.99615.94</a></span></p>
<p>4. Bruce D. Lucas and Takeo Kanade. 1981. An iterative image registration technique with an application to stereo vision. In Proceedings of the 7th international joint conference on Artificial intelligence - Volume 2 (IJCAI'81), Vol. 2. Morgan Kaufmann Publishers Inc., San Francisco, CA, USA, 674-679.</p>
<p>5. Greg Ward. Fast, robust image registration for compositing high dynamic range photographcs from hand-held exposures. Journal of Graphics Tools, 8(2):17–30, 2003.</p>
<p>6. Bingjian Wang; Dongliang Wu; Wenzheng Xu; Quan Lu; Fan Li; Shangqian Liu; Guowang Gao; Rui Lai; , "A new image registration method for infrared images and visible images," Image and Signal Processing (CISP), 2010 3rd International Congress on , vol.4, no., pp.1745-1749, 16-18 Oct. 2010</p>
<p>7. Barbara Zitová, Jan Flusser, Image registration methods: a survey, Image and Vision Computing, Volume 21, Issue 11, October 2003, Pages 977-1000, ISSN 0262-8856, 10.1016/S0262-8856(03)00137-9.</p>
<p><span>8.</span> Herbert Bay, Andreas Ess, Tinne Tuytelaars, and Luc Van Gool. 2008. Speeded-Up Robust Features (SURF). Comput. Vis. Image Underst. 110, 3 (June 2008), 346-359. DOI=10.1016/j.cviu.2007.09.014 <a href="http://dx.doi.org/10.1016/j.cviu.2007.09.014" target="_blank">http://dx.doi.org/10.1016/j.cviu.2007.09.014</a></p>
<p><span>9. </span>Feng Zhao; Qingming Huang; Wen Gao; , "Image Matching by Normalized Cross-Correlation," Acoustics, Speech and Signal Processing, 2006. ICASSP 2006 Proceedings. 2006 IEEE International Conference on , vol.2, no., pp.II, 14-19 May 2006</p>
<p>10. Shun'ichi Kaneko, Yutaka Satoh, Satoru Igarashi, Using selective correlation coefficient for robust image registration, Pattern Recognition, Volume 36, Issue 5, May 2003, Pages 1165-1173, ISSN 0031-3203, 10.1016/S0031-3203(02)00081-X.</p>
<p>11. Martin A. Fischler and Robert C. Bolles. 1981. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Commun. ACM 24, 6 (June 1981), 381-395. DOI=10.1145/358669.358692 <span><a href="http://doi.acm.org/10.1145/358669.358692" target="_blank">http://doi.acm.org/10.1145/358669.358692</a></span></p>
<p>12. Winder, S. & Brown, M. Picking the best DAISY. IEEE Conference on Computer Vision and Pattern Recognition (2009) 0, 178-185 (2009).<br clear="all"></p><br></div><br>
</div><br></div></body></html>