A New Cross-Triangular Search Algorithm for Fast Block Motion Estimation
碩士 === 國立中興大學 === 電機工程學系 === 93 === Block-matching motion estimation plays an important role in real-time video compression process. In block-matching motion estimation, search patterns with different shapes, sizes or the center-biased properties of motion vector distribution have a significant impa...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
2005
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/61524120004359697824 |
| Summary: | 碩士 === 國立中興大學 === 電機工程學系 === 93 === Block-matching motion estimation plays an important role in real-time video compression process. In block-matching motion estimation, search patterns with different shapes, sizes or the center-biased properties of motion vector distribution have a significant impact on the searching speed and quality of performance. Using this characteristic, a new cross-triangular search (CTS) algorithm is developed. The CTS algorithm essentially employs small cross-search pattern (SCSP) and two different kinds of triangular-search patterns (TSP) consecutively in the searching process. The TSP used in the CTS is different from the hexagonal-search pattern (HSP) used in the cross-diamond-hexagonal search (CDHS) algorithm. The simulation results show the CTS algorithm can have 1.1 times faster than using the CDHS algorithm.
|
|---|