Alignment_journal_version

Alignment_journal_version - Alignment of Non-Overlapping...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Alignment of Non-Overlapping Sequences Yaron Caspi Michal Irani Dept. of Computer Science and Applied Math The Weizmann Institute of Science 76100 Rehovot, Israel PLEASE PRINT IN COLOR Abstract This paper shows how two image sequences that have no spatial overlap between their fields of view can be aligned both in time and in space. Such alignment is possible when the two cameras are attached closely together and are moved jointly in space. The common motion induces “similar” changes over time within the two sequences. This correlated temporal behavior, is used to recover the spatial and temporal transformations between the two se- quences. The requirement of “coherent appearance” in standard image alignment techniques is therefore replaced by “coherent temporal behavior”, which is often easier to satisfy. This approach to alignment can be used not only for aligning non-overlapping sequences, but also for handling other cases that are inherently difficult for standard image alignment techniques. We demonstrate applications of this approach to three real-world problems: (i) alignment of non-overlapping sequences for generating wide-screen movies, (ii) alignment of images (sequences) obtained at significantly different zooms, for surveillance applications, and, (iii) multi-sensor image alignment for multi-sensor fusion. 1 Introduction The problem of image alignment (or registration) has been extensively researched, and successful approaches have been developed for solving this problem. Some of these approaches are based on matching extracted local image features, other approaches are based on directly matching image intensities. A review of some of these methods A shorter version of this paper appeared in ICCV 2001 [6].
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
can be found in [22] and [14]. However, all these approaches share one basic assumption: that there is sufficient overlap between the two images to allow extraction of common image properties, namely, that there is sufficient “similarity” between the two images (“Similarity” of images is used here in the broadest sense. It could range from gray-level similarity, to feature similarity, to similarity of frequencies, and all the way to statistical similarity such as mutual information [24]). In this paper the following question is addressed: Can two images be aligned when there is very little similarity between them, or even more extremely, when there is no spatial overlap at all between the two images? When dealing with individual images, the answer tends to be “No”. However, this is not the case when dealing with image sequences. An image sequence contains much more information than any individual frame does. In particular, temporal changes (such as dynamic changes in the scene, or the induced image motion) are encoded between video frames, but do not appear in any individual frame. Such information can form a powerful cue for alignment of two (or more) sequences. Caspi and Irani [5] and Stein [21] have illustrated an applicability of such an approach
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 06/13/2011 for the course CAP 6412 taught by Professor Staff during the Spring '08 term at University of Central Florida.

Page1 / 19

Alignment_journal_version - Alignment of Non-Overlapping...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online