Spatio-temporal segmentation based on motion and static segmentation

Dufaux, Frédéric; Moscheni, F.; Lippman, Andrew

doi:10.1109/icip.1995.529707

Cited by 61 publications

(26 citation statements)

References 12 publications

(1 reference statement)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Under the dominant-motion method (e.g., [4], [12], [23]), a single parametric motion is robustly fitted to all the pixels and then regions which agree with this motion are segmented as one layer and the process repeated on the rest. Alternatively, a different motion may be fitted to each region and then some clustering performed in parameter space to group regions with similar motions [24], [25], [26], [27], [28], [29]. The EM algorithm is also a good choice when faced with this type of estimation problem [19].…”

Section: Review Of Previous Workmentioning

confidence: 99%

Layered motion segmentation and depth ordering by tracking edges

Smith

Drummond

Cipolla

2004

IEEE Trans. Pattern Anal. Machine Intell.

View full text Add to dashboard Cite

Abstract-This paper presents a new Bayesian framework for motion segmentation-dividing a frame from an image sequence into layers representing different moving objects-by tracking edges between frames. Edges are found using the Canny edge detector, and the Expectation-Maximization algorithm is then used to fit motion models to these edges and also to calculate the probabilities of the edges obeying each motion model. The edges are also used to segment the image into regions of similar color. The most likely labeling for these regions is then calculated by using the edge probabilities, in association with a Markov Random Field-style prior. The identification of the relative depth ordering of the different motion layers is also determined, as an integral part of the process. An efficient implementation of this framework is presented for segmenting two motions (foreground and background) using two frames. It is then demonstrated how, by tracking the edges into further frames, the probabilities may be accumulated to provide an even more accurate and robust estimate, and segment an entire sequence. Further extensions are then presented to address the segmentation of more than two motions. Here, a hierarchical method of initializing the Expectation-Maximization algorithm is described, and it is demonstrated that the Minimum Description Length principle may be used to automatically select the best number of motion layers. The results from over 30 sequences (demonstrating both two and three motions) are presented and discussed.

show abstract

Section: Review Of Previous Workmentioning

confidence: 99%

Layered motion segmentation and depth ordering by tracking edges

Smith

Drummond

Cipolla

2004

IEEE Trans. Pattern Anal. Machine Intell.

View full text Add to dashboard Cite

show abstract

“…For example, segmentation plays an important role in the field of video object extraction [1], [2], [3]. Since homogeneous regions correspondent to meaningful objects (which are mostly inhomogeneous), many of the video object extraction algorithms first partition the image into homogeneous regions, and then, in order to extract the moving object, the regions are merged according to temporal information of the sequence.…”

Section: Introductionmentioning

confidence: 99%

Color image segmentation based on adaptive local thresholds

Navon

Miller

Averbuch

2005

Image and Vision Computing

142

View full text Add to dashboard Cite

The goal of still color image segmentation is to divide the image into homogeneous regions.Object extraction, object recognition and object-based compression are typical applications that use still segmentation as a low-level image processing. In this paper we present a new method for color image segmentation. The proposed algorithm divides the image into homogeneous regions by local thresholds. The number of thresholds and their values are adaptively derived by an automatic process, where local information is taken into consideration. First, the watershed algorithm is applied. Its results are used as an initialization for the next step, which is iterative merging process. During the iterative process regions are merged and local thresholds are derived. The thresholds are determined one-by-one at different times during the merging process. Every threshold is calculated by local information on any region and its surroundings. Any statistical information on the input images is not given. The algorithm is found to be reliable and robust to different kind of images.

show abstract

“…Various segmentation approaches have been investigated and most of them are based on examining the images in pixel domain [1][2][3][4][5][6][7]. Lucchese and Mitra [3] used 2D k-means clustering using color information, and then associating these clusters with appropriate luminance values, using 1D kmeans algorithm.…”

Section: Introductionmentioning

confidence: 99%

“…Their algorithm consists of three steps, region simplification, region growing, and motion-based region fusion. Dufaux et al [7] used spatiotemporal segmentation algorithm based on luminance information and motion parameters. The luminance is filtered by morphological operator, and then clustered using k-means algorithm.…”

Section: Introductionmentioning

confidence: 99%

Fast object detection and segmentation in MPEG compressed domain

Sukmarg

Rao

2000 TENCON Proceedings. Intelligent Systems and Technologies for the New Millennium (Cat. No.00CH37119)

View full text Add to dashboard Cite

In this paper, we present a fast algorithm for object detection and segmentation in MPEG compressed domain using color clustering, region merging based on spatiotemporal similarities, background/foreground classification, and pixel edge extraction. The features extracted from the blocks of segmented object in compressed domain can be used for fast object tracking and indexing at low level. Moreover, those blocks can be decompressed to obtain details of a specific object in pixel domain and can be used for high level indexing. By using our algorithm, we can reduce the amount of the information needed to be processed, and therefore, save the computational time, and increase the processing speed. Also we need to perform inverse DCT on only some parts of the image.

show abstract

Spatio-temporal segmentation based on motion and static segmentation

Cited by 61 publications

References 12 publications

Layered motion segmentation and depth ordering by tracking edges

Layered motion segmentation and depth ordering by tracking edges

Color image segmentation based on adaptive local thresholds

Fast object detection and segmentation in MPEG compressed domain

Contact Info

Product

Resources

About