Computationally Scalable Partial Distance Based Fast Search Motion Estimation

In this paper, we present a class of algorithms that use a partial distance metric to speedup the motion estimation process. The partial distance metric is used within the motion search to eliminate unlikely candidates through a thresholding process that enables computation scalability. We also propose a multiresolution variant which further reduces the complexity. Our results are comparable to state of the art approaches present a regular structure and are computationally scalable.

• Type: Conference
• Year: 2000
• Where: ICIP
• Authors: Krisda Lengwehasatit, Antonio Ortega
• Keywords: ICIP 2000, Image Processing, Motion Estimation Process, Partial Distance Metric, Thresholding Process