C. Stiller and J. Konrad. Estimating motion in image sequences: A tutorial on modeling and computation of 2D motion. IEEE Signal Processing, 16, 1999.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....4 describes the parallel platform. Section 5 includes experimental results, while the last section concludes the paper. 2. MOTION ESTIMATION ALGORITHMS A number of very different motion estimation algorithms have been proposed in the literature. Detailed reviews are given by [1] 4] 6] 9] [18], 14] 19] 20] These algorithms have been developed for various applications such as image sequence analysis, machine vision, robotics, image sequence restoration or image sequence coding. Block based motion estimation (ME) techniques are based on the minimization of a disparity measure. In ....

C. Stiller and J. Konrad, " Estimating motion in image sequences: a tutorial on modeling and computation of 2D motion," in IEEE Signal Processing Magazine, pp. 70-98, July 1999.

....body of the car. A situation like this requires a good modeling approach and in this paper we propose one that is established in probabilistic framework. Probabilistic interpretation of global cost function for motion estimation, known as MAP estimation, has two models, i.e. observation and prior [1]. The prior model includes a smoothness constraint or displacement model and a line field model for discontinuity constraint. The observation model directly analyzes data[2] Heitz and Bouthemy [3] have built observation model that includes intensity edge detector and validity test of the ....

....that by Dubois and Konrad [7] However, the region of support of the observation model d U in this paper, is not in a frame but rather in a neighborhood system for matching purposes. Ordinarily, the block matching writes motion vector for a whole block and it is known as the region of support in [1]. Our novel algorithm uses an observation model accompanied by the adaptive block size order of neighborhood system, and the result is applied for one motion vector in the middle of a block. In the energy function in equation (3) l U is a displacement field model, and h U is a block size model. ....

C. Stiller and J. Konrad, " Estimating Motion in Image Sequences, A tutorial on modeling and computation of 2D motion," IEEE Signal Process. Magazine, pp. 70 - 98, July 1999.

....Fleet and Jepson detect neighborhoods around phase singularities, requiring spatial derivatives, which cannot be computed purely locally. Finally, our technique assumes the optical flow field to be uniform within the spatial extent of our filter pairs (translational model; for an overview, see [22]) A more sophisticated scheme, such as the affine model of motion, could be used, which would result in a quadratic, rather than a linear regression of the phase over time. The constraints corresponding to the resulting component motion fields would, however, become more complicated. All these ....

C. Stiller, and J. Konrad, "Estimating Motion in Image Sequences, A tutorial on modeling and computation of 2D motion," IEEE Signal Process. Mag., vol. 16, pp. 70--91, 1999.

No context found.

C. Stiller and J. Konrad. Estimating motion in image sequences: A tutorial on modeling and computation of 2D motion. IEEE Signal Processing, 16, 1999.

No context found.

C. Stiller and J. Konrad. Estimating motion in image sequences: A tutorial on modeling and computation of 2D motion. IEEE Signal Process, 16:70--91, 1999.

No context found.

Stiller C. & Konrad J. (1999) Estimating Motion in Image Sequences---A tutorial on modeling and computation of 2D motion. IEEE Signal Processing Magazine 16 (4), pp. 70--91.

No context found.

C. Stiller and J. Konrad. Estimating motion in image sequences: A tutorial on modeling and computation of 2D motion. IEEE Signal Process, 16:70--91, 1999.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC