Spatial Prediction&Gradient Descent-based Fast Motion Estimation Algorithm Using Square-Diamond Search Strategy

Hua Yongqiang,Zhao Chunhui,Cheng Yongmei

DOI: https://doi.org/10.16526/j.cnki.11-4762/tp.2008.07.019

2008-01-01

Abstract:To improve the accuracy and enhance the speed of block-matching-based motion estimation,a robust square-diamond search algorithm depending on spatial prediction and gradient descent(PGDS) is presented in this paper.The key of this algorithm is judging the correlation between current block and adjacent blocks.If there were correlation between them,we compute the vector of motion prediction and search for the motion vector based on prediction using small diamond search pattern(SDSP).Otherwise,we search for motion vector from the center point using gradient descent-based square-diamond search algorithm(GDS).Experimental results showed that the proposed algorithm improves the accuracy and enhances the speed of motion estimation evidently.At the same time,this algorithm is more robust than TSS or DS.