A Stochastic Framework for Optimal Key Frame Extraction from MPEG Video Databases

Avrithis, Yannis; Doulamis, Anastasios; Kollias, Stefanos

doi:10.1006/cviu.1999.0761

Cited by 87 publications

(79 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In other words, histogram does not internment spatial patterns that are imperative to describe image content [17]. To overcome this restriction, we use histograms of numerous spectral components of an image as the signature, which are used in texture analysis and synthesis [19].…”

Section: Introductionmentioning

confidence: 99%

A Content Based Video Retrieval Analysis System with Extensive Features by Using Kullback-Leibler

Priya¹,

Shanmugam²,

Baskaran³

2014

IJCIS

View full text Add to dashboard Cite

Content-based video retrieval systems have shown great potential in supporting decision making in clinical activities, teaching, and biological research. In content-based video retrieval, feature combination plays a key role. As a result content-based retrieval of all different type video data turns out to be a challenging and vigorous problem. This paper presents an effective content based video retrieval system, which recognizes and retrieves videos with three different types of visual effects. The raw video information is divided into shots and also the object feature, movement feature and also the occlusion options are extracted from these shots and also the feature library is used for the storage method of those options. Advanced on, the Kullback-Leibler distance is computed among the options of the feature library and also the options of the question clip that's extracted within the similar manner. The results show that it is possible to improve a system for content-based video retrieval by using Kullback-Leibler distance model, which takes careful consideration of the structure and distribution of visual features. Hence the final results with the aid of the Kullback-Leibler distance the similar videos are extracted from the collection of videos based on the given query video clip in an effective manner.

show abstract

Section: Introductionmentioning

confidence: 99%

A Content Based Video Retrieval Analysis System with Extensive Features by Using Kullback-Leibler

Priya¹,

Shanmugam²,

Baskaran³

2014

IJCIS

View full text Add to dashboard Cite

show abstract

“…Towards this direction the conventional M-RSST segmentation algorithm is incorporated, where results of the lowest resolution level are not just propagated but also adjusted at higher levels [16]. This is essential as color segments accurately detect the boundaries of video objects.…”

Section: Segmentation Results Of the Cfcs Approachmentioning

confidence: 99%

“…In this paper a modified version of the multiresolution Recursive Shortest Spanning Tree segmentation algorithm (M-RSST) [16], is incorporated for depth segmentation. The M-RSST recursively applies the RSST to images of increasing resolution and the RSST iteration phase terminates when either the total number of segments or the minimum link weight reaches a target threshold.…”

Section: Occlusion Detection/compensation and Depth Field Segmentationmentioning

confidence: 99%

“…The first step of the proposed scheme includes partitioning of one of the image channels (left or right) into color segments. Now the conventional M-RSST algorithm is incorporated, where initial segmentation at the lowest resolution level is not just propagated but it is also adjusted at higher levels [16]. After producing a color segments map, color segments can be fused according to depth similarity criteria to provide the video objects.…”

Section: The "Constrained Fusion Of Color Segments" (Cfcs) Approachmentioning

confidence: 99%

“…However as the final polygon (initial active contour) depends on the order that points are discarded and in order to keep points belonging to different areas, the mean color difference MC between triplets of adjacent points guides the process. Considering that P i-1 , P i and P i+1 are three neighboring points and the image is in RGB format, the mean color difference MC i for point P i is defined as: (16) In the proposed scheme an iterative merging technique is incorporated, where starting from polygon P 1 P 2 …P N P 1 and by discarding points according to color homogeneity and shape distance criteria, the initial active contour is produced. More specifically let SD= {P 1 , P 2 , …, P N } denote the ordered set of points comprising the nodes of the constructed polygon, like the one depicted in Figure 4.…”

Section: Initial Points Selectionmentioning

confidence: 99%

See 2 more Smart Citations

Unsupervised segmentation of stereoscopic video objects: investigation of two depth-based approaches

Ntalianis

Doulamis

et al.

2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)

View full text Add to dashboard Cite

In this paper two efficient unsupervised video object segmentation approaches are proposed and thoroughly compared in terms of computational cost and quality of segmentation results. Both methods are based on the exploitation of depth information, estimated for stereoscopic pairs of frames. In particular, in both schemes an occlusion compensated disparity field is initially computed and a depth map is generated. Then a depth segments map is produced by incorporating a modified version of the multiresolution Recursive Shortest Spanning Tree segmentation algorithm (M-RSST). Next considering the first "Constrained Fusion

show abstract

A new paradigm for analysis of MPEG compressed videos

Farag

Abdel-Wahab

2002

Journal of Network and Computer Applications

View full text Add to dashboard Cite

A Stochastic Framework for Optimal Key Frame Extraction from MPEG Video Databases

Cited by 87 publications

References 31 publications

A Content Based Video Retrieval Analysis System with Extensive Features by Using Kullback-Leibler

A Content Based Video Retrieval Analysis System with Extensive Features by Using Kullback-Leibler

Unsupervised segmentation of stereoscopic video objects: investigation of two depth-based approaches

A new paradigm for analysis of MPEG compressed videos

Contact Info

Product

Resources

About