Browse Prior Art Database

DETECTING TIME-VARYING CORNERS

IP.com Disclosure Number: IPCOM000128462D
Original Publication Date: 1983-Nov-01
Included in the Prior Art Database: 2005-Sep-16
Document File: 9 page(s) / 101K

Publishing Venue

Software Patent Institute

Related People

Shah, Mubarak A.: AUTHOR [+4]

Abstract

The algorithms for structure from motion require solution of the correspondence problem. By detecting only time-varying tokens, the problem may be significantly simplified. In this paper, a time varying corner detector is described which is based on the and operation between the cornerness and the temporal derivative. It is shown that the corner detectors by Zuniga and Haralick, Kitchen and Rosenfeld, and Dreschler and Nagel are equivalent. In our time-varying corner detector, we use Zuniga and Haralick corner detector for finding the cornerness at a point and use absolute value of difference in intensity at a point to approximate the temporal derivative. The results of the time varying corner detector for the the real scenes and the synthetic images with random background and random object are shown .

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 12% of the total text.

Page 1 of 9

THIS DOCUMENT IS AN APPROXIMATE REPRESENTATION OF THE ORIGINAL.

DETECTING TIME-VARYING CORNERS

Mubarak A. Shah and Ramesh Jain

CRL-TR-34-83
THE UNIVERSITY OF MICHIGAN COMPUTING RESEARCH LABORATORY NOVEMBER 1983

Room 1079, East Engineering Building

Ann Arbor, Michigan 48109
USA
Tel: (313) 763 8000

DETECTING TIME-VARYING CORNERS

Mubarak A. Shah and Ramesh Jain

Department of Electrical and Computer Engineering

University of Michigan
Ann Arbor, MI 48109

ABSTRACT

The algorithms for structure from motion require solution of the correspondence problem. By detecting only time-varying tokens, the problem may be significantly simplified. In this paper, a time varying corner detector is described which is based on the and operation between the cornerness and the temporal derivative. It is shown that the corner detectors by Zuniga and Haralick, Kitchen and Rosenfeld, and Dreschler and Nagel are equivalent. In our time-varying corner detector, we use Zuniga and Haralick corner detector for finding the cornerness at a point and use absolute value of difference in intensity at a point to approximate the temporal derivative. The results of the time varying corner detector for the the real scenes and the synthetic images with random background and random object are shown .

1. Introduction

The time varying features play an important role in dynamic scene analysis. The biological systems are capable of detecting features in the environment which are changing with the time. The features may be varying either due to the motion of an object or the observer. The human vision system utilizes both of these means to perceive the world. Certain objects become clear when we move our eyes, head or the body 1. Similarly the perception is improved when the objects move.

It is well known that the structure of an object can be determined from its motion. In various approaches to structure from motion [ U1179, RoA80, TsH82 ] , correespondence of N points in M frames is required. The correspondence problem is solved by detecting interesting point' in frames. Corners are considered good candidates for establishing correspondence [ DrN79 ] . In the proposed approaches for structure from motion the corners are detected in each frame. Since each frame may contain many stationary corners, in addition to a few moving corners, the

1 Gibson ,J. J. Ecological Approach to Visual Perception Hougton Mifflen, Boston ,1979.

University of Michigan Computing Research Laboratory Page 1 Nov 01, 1983

Page 2 of 9

DETECTING TIME-VARYING CORNERS

correspndence becomes more difficult. If only moving corners could be detected, then the matching of corners from frame to frame may become more tractable and computationally efficient. Our efforts to use corners in each frame to determine motion parameters 2 showed the difficulty in working with corner detectors and encouraged us to investigate possibility of detecting only time-varying corners.

Another application of time-varying corners may b...