A New Motion Estimation Method Using Frequency Components

✍ Scribed by Yung-Ming Chou; Hsueh-Ming Hang

• Publisher: Elsevier Science
• Year: 1997
• Tongue: English
• Weight: 839 KB
• Volume: 8
• Category: Article
• ISSN: 1047-3203
• DOI: 10.1006/jvci.1997.0343

✦ Synopsis

on the assumption that the image intensity can be viewed as an analytic function in spatial and temporal domains.

Motion estimation techniques are widely used in today's video processing systems. The most frequently used techniques It was first proposed by Cafforio and Rocca [8], and later are the block matching method and the differential method. Netravali and Robbins developed an iterative algorithm, In this paper, we have studied this topic from a viewpoint the so-called pel-recursive method [9]. The optical flow different from the above to explore the fundamental limits and method [10] in computer vision is much like the pel-retradeoffs in image motion estimation. The underlying principles cursive scheme even though they were derived from differbehind two conflict requirements in motion estimation, accuent bases. There are many refined versions of the pelracy and ambiguity, become clear when they are analyzed using recursive however, and optical flow methods [11][12][13][14][15][16][17][18]. The this tool-frequency component analysis. This analysis also Fourier method used for motion estimation, however, is suggests new motion estimation algorithms and ways to imnot as popular as these two approaches.

prove the existing algorithms. The so-called frequency component motion estimation algorithm is thus proposed. Compared Phase correlation [19] is the most well-known method to the conventional block matching and phase correlation algoin this class that utilizes phase information of frequency rithms, this approach provides more reliable displacement esticomponents in estimating the motion vectors. Thomas [20, mates particularly for the noisy pictures.