Global motion estimation from coarsely sampled motion vector field and the applications

Su, Yeping; Sun, Ming–Ting; Hsu, Vincent

doi:10.1109/iscas.2003.1206052

Cited by 7 publications

(7 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The same four sets of global motion parameters as in [2,4], which are shown in Table 1, are used. The synthetic motion vector fields of these four motion models are illustrated in Fig.…”

Section: Resultsmentioning

confidence: 99%

“…Four popular motion models are: translational (2 parameters), geometric (4 parameters), affine (6 parameters) and perspective (8 parameters). GME can be performed in either the pixel domain [1] or the compressed domain [2][3][4] by solving a set of linear equations with the help of feature points correspondence or a motion vector field. An evaluation of pixel based and motion vector based GME methods can be found in [5].…”

Section: Introductionmentioning

confidence: 99%

“…To estimate the global motion from a coarsely sampled motion vector field, the Newton-Raphson gradient descent method is used in [2] to minimize the fitting error between the input motion vectors and those generated from the estimated motion model. In [6], the least square solution for GME is obtained using an M-Estimator.…”

Section: Introductionmentioning

confidence: 99%

“…Several methods have been reported to remove the outliers that fit poorly into the global motion model. In [2], an iterative approach with fairly high complexity is used. In [3], the outliers are detected by examining the magnitude difference between a vector and its eight neighbors.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Efficient motion vector outlier removal for global motion estimation

Dinh

Lee

2011

2011 IEEE International Conference on Multimedia and Expo

View full text Add to dashboard Cite

Because motion vector based global motion estimation methods have much lower complexity than pixel based ones, they are widely used in the compressed domain to estimate the camera motion in video sequences. However, the accuracy of these motion vector based methods largely depends on the quality of the input motion vector field. In real applications, many outlier motion vectors are present because of noise or foreground objects. In this paper, a novel tensor voting based motion vector outlier removal method is proposed to improve the quality of the input motion vector field. First, motion vectors are encoded by second order tensors. A 2-D voting process is then used to smooth the motion vector field. Finally, the smoothed motion vector field is compared to the input one to detect outliers. The experimental results on synthetic and real data show the effectiveness of the proposed method.

show abstract

“…The same four sets of global motion parameters as in [2,4], which are shown in Table 1, are used. The synthetic motion vector fields of these four motion models are illustrated in Fig.…”

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Efficient motion vector outlier removal for global motion estimation

Dinh

Lee

2011

2011 IEEE International Conference on Multimedia and Expo

View full text Add to dashboard Cite

show abstract

“…The problem with this error discarding procedure is that it accepts many foreground pixels and avoids background pixels, thus degrading the estimation process. Some other techniques that utilize robust estimators to deal with the local motion problem are discussed in [13,14].…”

Section: Introductionmentioning

confidence: 99%