一種基于時空相關性的運動估計算法
A NEW MOTION ESTIMATION ALGORITHM BASED ON SPATIO-TEMPORAL COHERENCE
-
摘要: 研究了一種基于時空相關性的運動估計方法。提出在塊匹配法中,對于運動相關性好的塊根據(jù)相鄰塊或前一幀圖像的運動預測出它的初始運動矢量,然后在小范圍內(nèi)搜索,做運動矢量的優(yōu)化。對傳統(tǒng)的塊匹配運動估計,給出了提高運動矢量一致性的修正準則。實驗結果表明,本文算法所得的運動矢量一致性遠高于傳統(tǒng)塊匹配法,同時能保持和全局搜索法相媲美的預測質(zhì)量,而運算時間則有較大幅度縮短。
-
關鍵詞:
- 運動估計; 塊匹配法; 時空相關性
Abstract: A new motion estimation algorithm based on spatio-temporal cohenence is presented. In this proposed algorithm, an initial motion vector of the block which has good coherence with its neighbors or that of the previous frame is directly predicated. A more precise motin vector is obtained based on the initial one by full-searching block matching algorithm (FBMA) in a small range. As to blocks with poor coherence, FBMA with ordinary range is used, and their motion vectors are at last adjusted by the coherence criterion given in the paper. Experiments show satisfying results can be reached with the proposed algorithm. -
Dufaux F, Moscheni F. Motion estimation techniques for digital TV: a review and a new contri-[2]bution, Proc. IEEE; 1995, 83(6): 858-876.[3]Haskeu B G, et al. Image and Video Coding Emerging Standards and beyond, IEEE Trans on-CAS VT, 1998; 8(7) 814-837.[4]Hangu Yeo.[J].Yu Hen Hu. A motion estimation and image segmentation technique based on the variable block size. Proc. ICASSP97, Munich, Germany.3137,1997:-[5]Bierling M. Displacement estimation by hierarchical block matching SPIE Vo1.1001, VCIP88, 942-951.[6]Yuichiro Nakaya, Hiroshi Harashima. Motion compensation based on spatial tranformations IEEE Trans on CASVT, 1994, CASVT-4(3):339-356.[7][6][8]Deihl N. Object-oriented motion estimation[9]and segmentation in image sequences. Signal Processing: Image Communication, 1991,3(1): 23-56.[10]Aizawa K, Huang T S. Model-based image coding advanced video coding techniques for very low[11]bit-rate applications. Proc. IEEE, 1995, 83(2) 259-271.[12]薛向陽, 吳立德. 基于運動矢量插值的運動補償算法. 電子科學學刊.1997, 19(4): 483-488. -
計量
- 文章訪問數(shù): 2039
- HTML全文瀏覽量: 113
- PDF下載量: 511
- 被引次數(shù): 0
下載:
