Motion estimation using integral projection
An image sequence comprises at least a first image and a second image. A motion vector (MV(1)) is associated to these first and second images and is defined in a coordinate system having at least a first direction and a second direction, the second direction being different from said first direction...
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Patent |
| Sprache: | eng ; fre ; ger |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Zusammenfassung: | An image sequence comprises at least a first image and a second image. A motion vector (MV(1)) is associated to these first and second images and is defined in a coordinate system having at least a first direction and a second direction, the second direction being different from said first direction. This motion vector has a first previous component value (MV D1 (0)) in the first direction. A method processes both images in order to determine an estimate of the motion vector based on the first previous component value. To this purpose, the method defines (201) a first image portion in the first image and a second image portion in the second image based on the first previous component value of the motion vector. Then, the method determines (202) a first characteristics vector for the first image in the first image portion and a second characteristics vector for the second image in the second image portion. The first and second characteristics vectors are compared to estimate (203) a first shift value between the first and second images. At last, a second current component value and a first current component value are obtained and the motion vector is defined by these first and second current component values. |
|---|