&lt;title&gt;Scene change detection and feature extraction for MPEG-4 sequences&lt;/title&gt;

Divakaran, Ajay; Ito, Hiroshi; Sun, Huifang; Poon, T.C.

doi:10.1117/12.333874

Cited by 6 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous work in feature extraction for indexing in the compressed domain falls in the following categories: In our previous work [3] we have described a computationally simple technique that combines DC images [9} with bit allocation [4] information so as to simply and accurately detect abrupt scene changes. This motivated us to investigate techniques that make use of bit allocation information to extract features for indexing [1] since the spatial distribution of the bits spent on motion is an indication of the spatial distribution of the spatio-temporal complexity in the object/frame.…”

Section: Motivationmentioning

confidence: 99%

<title>Descriptor for spatial distribution of motion activity for compressed video</title>

Divakaran

Sun

1999

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

In this paper we present a new descriptor for spatial distribution of motion activity in video sequences. We use the magnitude of the motion vectors as a measure of the intensity of motion activity in a macro-block. We construct a matrix Cmv consisting of the magnitudes of the motion vector for each macro-block of a given P frame. We compute the average magnitude of the motion vector per macro-block, Cavg, and then use Cavg as a threshold on the matrix C by setting the elements of C that are less than Cavg to zero. We classify the runs of zeroes into three categories based on length, and count the number of runs of each category in the matrix C. Our activity descriptor for a frame thus consists of four parameters viz. the average magnitude of the motion vectors and the numbers of runs of short, medium and long length. Since the feature extraction is in the compressed domain and simple, it is extremely fast. We have tested it on the MPEG-7 test content set, which consists of approximately 14 hours of MPEG-i encoded video content of different kinds. We find that our descriptor enables fast and accurate indexing of video. It is robust to noise and changes in encoding parameters such as frame size, frame rate, encoding bit rate, encoding format etc. It is a low-level non-semantic descriptor that gives semantic matches within the same program, and is thus very suitable for applications such as video program browsing. We also find that indirect and computationally simpler measures of the magnitude of the motion vectors such as bits taken to encode the motion vectors, though less effective, also can be used in our mn-length framework.

show abstract

Section: Motivationmentioning

confidence: 99%

<title>Descriptor for spatial distribution of motion activity for compressed video</title>

Divakaran

Sun

1999

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

show abstract

“…Divakaran et al 14,15 use features such as mean, standard deviation and directionality of the motion vectors over the video shots, exploiting the motion vector information included in the MPEG-2 binary streams. Fablet et al 16 propose a global interpretation of the dynamic content of video shots in terms of spatio-temporal distributions modeled by causal Gibbs models.…”

Section: Introductionmentioning

confidence: 99%

<title>Parametric motion models for video content description within the MPEG-7 framework</title>

Zaharia¹,

Prêteux²

2001

SPIE Proceedings

View full text Add to dashboard Cite

This paper addresses the issue of motion-based description of video contents within the framework of the future MPEG-7 standard. Motion is a highly relevant feature related to the spatio-temporal structure of a video and therefore should play a central role within any attempt of content-based video description. We show that parametric motion models provide a powerful tool for characterizing motion of arbitrary (foreground or background) video objects within an unified mathematical framework. The parametric object motion descriptor (POMD) principle consists of describing the motion of objects in video sequences as a 2D geometric transform. The motion models considered in this paper are those currently adopted in the future MPEG-7 standard: constant, simplified scale/rotation, affine, planar perspective and parabolic. The descriptor is defined as the vector of parameters associated with the geometric transform providing thus a very compact representation of the video object motion. Several similarity measures between motion models are considered, including simple L 1 or L 2 distances in the parameter space, and more elaborate similarity measures based upon distance functions directly related to the velocity fields. The experimental results, carried out upon the data sets created by the authors for evaluating the motion descriptors within the MPEG-7 development process, are presented and discussed.

show abstract

“…In the compressed domain, the DC-image has been used widely in shotboundary detection and video segmentation due to its small size [10][11][12][13][14]. It was also utilized for keyframe extraction, instead of parsing the full frame [15][16][17], or even for video summarization purpose [18,19].…”

Section: Introductionmentioning

confidence: 99%

DC-Image for Real Time Compressed Video Matching

Bekhet

Ahmed

Hunter

2014

Transactions on Engineering Technologies

View full text Add to dashboard Cite

This chapter presents a suggested framework for video matching based on local features extracted directly from the DC-image of MPEG compressed videos, without full decompression. In addition, the relevant arguments and supporting evidences are discussed. Several local feature detectors will be examined to select the best for matching using the DC-image. Two experiments are carried to support the above. The first is comparing between the DC-image and I-frame, in terms of matching performance and computation complexity. The second experiment compares between using local features and global features regarding compressed video matching with respect to the DC-image. The results confirmed that the use of DC-image, despite its highly reduced size, is promising as it produces higher matching precision, compared to the full I-frame. Also, SIFT, as a local feature, outperforms most of the standard global features. On the other hand, its computation complexity is relatively higher, but it is still within the real-time margin which leaves a space for further optimizations that could be done to improve this computation complexity.

show abstract

<title>Scene change detection and feature extraction for MPEG-4 sequences</title>

Cited by 6 publications

References 0 publications

<title>Descriptor for spatial distribution of motion activity for compressed video</title>

<title>Descriptor for spatial distribution of motion activity for compressed video</title>

<title>Parametric motion models for video content description within the MPEG-7 framework</title>

DC-Image for Real Time Compressed Video Matching

Contact Info

Product

Resources

About