Title of Dissertation: View Synthesis from Image and Video for Object Recognition Applications Zhanfeng Yue, Doctor of Philosophy, 2007 Dissertation directed by: Professor Rama Chellappa Department of Electrical and Computer Engineering Object recognition is one of the most important and successful applications in computer vision community. The varying appearances of the test object due to diff...

Existing featureless image mosaicing techniques do not pay enough attention to the robustness of the image registration process, and are not able to combine multiple video sequences into an integrated panoramic view. These problems have certainly restricted applications of the existing methods for large-scale panorama composition, video content overview and information visualization. In this pa...

In this paper, a multi-view video coding method via generation of virtual picture sequences is proposed. Pictures are synthesized for the sake of better exploitation of the redundancies between neighbouring views in a multi-view sequence. Pictures are synthesized through a 3D warping method to estimate certain views in a multi-view set. Depth map and associated colour video sequences are used f...

In this paper we describe elements of an ubiquitous vision system a system that enables viewers to observe a remote dynamic scene from any viewing perspective at any time instant. This integrated framework has the potential to achieve realtime generation of many 3D virtual views concurrently with high-dynamic events in the scene, by not utilizing 3D model. The paper presents a brief overview of...

3D imaging and video technologies are of growing interest in recent times because of having potential applications in many fields such as robotics, visualization, 3DTV, autonomous vehicles, driver assistance, “flying eyes”, intelligent human-machine interfaces, and so forth. To advance those technologies, computer vision and pattern recognition, which originated separately, are today increasing...

iran’s industry has had a considerable number of difficulties to perform so well as was expected in the post – revolutionary era. unfavorable legal, political environments combined with eight –year imposed war may be singled out as the root causes of such performances. however, this paper argues that our industry’s lack of realization of its dire need to adapt to the emerging unprecedented envi...

This paper presents an approach for realistic virtual view generation using appearance clustering based local view transition model, with its target application on cross-pose face recognition. Previously, the traditional global pattern based view transition model (VTM) method was extended to its local version called LVTM, which learns the linear transformation of pixel values between frontal an...