Layered representation of motion video using robust maximum-likelihood estimation of mixture models and MDL encoding

Ayer, Steven K.; Sawhney, Harpreet

doi:10.1109/iccv.1995.466859

Cited by 205 publications

(168 citation statements)

References 14 publications

Supporting

Mentioning

168

Contrasting

Order By: Relevance

“…Finally, a fourth related category is composed of other methods, that are based on mix-ture models [2,21], MDL criterion principle [2,24], or the robust regression framework [3]. In [2], motion models and spatial supports of the mixture are simultaneously estimated as the two steps of an EM algorithm.…”

Section: Different Techniques Have Been Investigated For Motion Segmementioning

confidence: 99%

“…In [2], motion models and spatial supports of the mixture are simultaneously estimated as the two steps of an EM algorithm. One drawback of mixture-based methods, also true for clustering methods, is that they do not inherently incorporate spatial coherence in the estimation of the spatial supports.…”

Section: Different Techniques Have Been Investigated For Motion Segmementioning

confidence: 99%

“…The block effect observed with the more traditional schemes is much reduced with such a region-based approach. Secondly, it is useful to analyse and interpret the scene, as well as to extract the different moving parts visually important for the task at hand, sometimes referred to as layers [1,2]. The choice of motion as a cue for image segmentation allows us to deal with a small number of meaningful entities, as opposed for instance to a gray-level based segmentation approach, and to build a compact and structured representation of the spatio-temporal content of a video sequence.…”

Section: Introduction and Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Direct incremental model-based image motion segmentation for video analysis

Odobez

Bouthémy

1998

Signal Processing

View full text Add to dashboard Cite

Dynamic analysis of image sequences is an important task in object-oriented video applications. It often relies on the segmentation of each image of the sequence into region entities of apparent homogeneous motion. In this paper, we present an original motion segmentation algorithm based on 2D polynomial motion models, a multiresolution robust estimator to compute these motion models, and appropriate local observations supplying both motion relevant information and their reliability.Motion segmentation is formulated as a contextual statistical labeling problem exploiting multiscale Markov Random Field (MRF) models. One of its main features is that it avoids time consuming alternate iterations between motion model estimation and spatial support identification. An original detection step allows us to estimate and to update the number of required motion models, and thus to handle the appearance of new objects. Numerous experiments performed with real indoor and outdoor image sequences demonstrate the efficiency of the method.

show abstract

Section: Different Techniques Have Been Investigated For Motion Segmementioning

confidence: 99%

Section: Different Techniques Have Been Investigated For Motion Segmementioning

confidence: 99%

Section: Introduction and Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Direct incremental model-based image motion segmentation for video analysis

Odobez

Bouthémy

1998

Signal Processing

View full text Add to dashboard Cite

show abstract

“…Video segmentation, or layer extraction, is a classic inverse problem in computer vision that involves the extraction of foreground objects from a set of images [4,17,33]. In image segmentation the goal is to segment an image into spatially coherent regions, whereas in video segmentation the goal is segment the image into temporally coherent regions.…”

Section: Introductionmentioning

confidence: 99%

Automatic video segmentation using spatiotemporal T-junctions

Apostoloff¹,

Fitzgibbon²

2006

Procedings of the British Machine Vision Conference 2006

View full text Add to dashboard Cite

The problem of figure-ground segmentation is of great importance in both video editing and visual perception tasks. Classical video segmentation algorithms approach the problem from one of two perspectives. At one extreme, global approaches constrain the camera motion to simplify the image structure. At the other extreme, local approaches estimate motion in small image regions over a small number of frames and tend to produce noisy signals that are difficult to segment. With recent advances in image segmentation showing that sparse information is often sufficient for figureground segmentation it seems surprising then that with the extra temporal information of video, an unconstrained automatic figure-ground segmentation algorithm still eludes the research community. In this paper we present an automatic video segmentation algorithm that is intermediate between these two extremes and uses spatiotemporal features to regularize the segmentation. Detecting spatiotemporal T-junctions that indicate occlusion edges, we learn an occlusion edge model that is used within a colour contrast sensitive MRF to segment individual frames of a video sequence. T-junctions are learnt and classified using a support vector machine and a Gaussian mixture model is fitted to the (foreground, background) pixel pairs sampled from the detected T-junctions. Graph cut is then used to segment each frame of the video showing that sparse occlusion edge information can automatically initialize the video segmentation problem.

show abstract

“…[9] fits a mixture of parametric models by minimizing a Mumford-Shah-like cost functional. [3][4][5][6][7][8] fit a mixture of probabilistic models iteratively using the Expectation Maximization algorithm (EM). The drawback of such iterative approaches is that they are very sensitive to correct initialization and are computationally expensive.…”

Section: Introductionmentioning

confidence: 99%

Direct Segmentation of Multiple 2-D Motion Models of Different Types

Singaraju

Vidal

Dynamical Vision

View full text Add to dashboard Cite

Abstract. We propose a closed form solution for segmenting mixtures of 2-D translational and 2-D affine motion models directly from the image intensities. Our approach exploits the fact that the spatial-temporal image derivatives generated by a mixture of these motion models must satisfy a bi-homogeneous polynomial called the multibody brightness constancy constraint (MBCC). We show that the degrees of the MBCC are related to the number of motions models of each kind. Such degrees can be automatically computed using a one-dimensional search. We then demonstrate that a sub-matrix of the Hessian of the MBCC encodes information about the type of motion models. For instance, the matrix is rank-1 for 2-D translational models and rank-3 for 2-D affine models. Once the type of motion model has been identified, one can obtain the parameters of each type of motion model at every image measurement from the cross products of the derivatives of the MBCC. We then demonstrate that accounting for a 2-D translational motion model as a 2-D affine one would result in erroneous estimation of the motion models, thus motivating our aim to account for different types of motion models. We apply our method to segmenting various dynamic scenes.

show abstract

Layered representation of motion video using robust maximum-likelihood estimation of mixture models and MDL encoding

Cited by 205 publications

References 14 publications

Direct incremental model-based image motion segmentation for video analysis

Direct incremental model-based image motion segmentation for video analysis

Automatic video segmentation using spatiotemporal T-junctions

Direct Segmentation of Multiple 2-D Motion Models of Different Types

Contact Info

Product

Resources

About