一种高精度的压缩域视频目标分割算法

唐志峰; 王诗俊; 杨树元

doi:10.3724/SP.J.1146.2006.00644

一种高精度的压缩域视频目标分割算法

doi: 10.3724/SP.J.1146.2006.00644

计量
- 文章访问数: 3199
- HTML全文浏览量: 85
- PDF下载量: 1123
- 被引次数: 0
出版历程
- 收稿日期: 2006-05-15
- 修回日期: 2006-09-20
- 刊出日期: 2007-12-19

A High Precision Compressed Domain Approach for Video Object Segmentation

摘要

摘要: 该文提出了一种工作于MPEG压缩域的快速视频目标分割算法。该算法以从MPEG1/2码流中部分解码提取的特征为输入，提取P帧中的运动目标。针对一般的压缩域算法目标边界精度不高的特点，算法采用I帧和P帧中每个块的直流DCT系数和3个交流DCT系数，以及运动补偿信息，重建出P帧的原图像1/16大小的子图像，采用快速平均移聚类得到具有较高边界精度的亮度一致的区域；针对运动向量的噪声容易造成错误检测的缺点，算法结合聚类分析结果和运动块的分布，采用基于马尔可夫随机场的统计标号方法对目标和背景区域进行分类，得到每个P帧的目标掩模。该算法可以得到44子块的边界精度，对于CIF格式的码流，在Pentium IV 2GHz平台上可以达到每秒40帧的处理速度。
- 视频目标分割;压缩域;快速平均移聚类;马尔可夫场
Abstract: A fast video object segmentation method working in MPEG compressed domain is presented in this paper. Moving object masks in P frames are extracted by exploiting features obtained by partial decoding. To increase object boundary precision, for each P frame, a 1/16 sub image is constructed using DC and three AC coefficients, and motion compensation information, then a fast mean shift clustering algorithm is used to divide the image into regions with coherence luminance and obtain high precision region boundaries. For reducing the influence of motion vector noise, a MRF-based statistical labeling method is exploited to classify regions into two classes: moving object and background. The proposed algorithm can get a boundary precision of 44 sub-block with a high processing speed. For CIF video streams, the algorithm can run at a speed of 40 frames per second in a Pentium IV 2GHz platform.

HTML全文

参考文献(1)

[1] Megret R, and DeMenthon D. A survey of spatio-temporal grouping techniques. Technical report: University of Maryland, College Park; 2002. [2] Tsaig Y, and Averbuch A. Automatic segmentation of moving objects in video sequences: a region labeling approach[J].IEEE Trans. on Circuits and Systems for Video Technology.2002, 12(7):597-612 [3] Xiao J, and Shah M. Motion layer extraction in the presence of occlusion using graph cuts[J].IEEE Trans. on Pattern Analysis and Machine Intelligence.2005, 27(10):1644-1659 [4] Wang R, Zhang H J, and Zhang Y Q. A confidence measure based moving object extraction system built for compressed domain. IEEE International Symposium on Circuits and Systems, Geneva, Switzerland, 2000, 5: 21-24. [5] Porikli F M. Real-time video object segmentation for MPEG-encoded video sequences. SPIE Conference on Real-Time Imaging VIII, 2004, 5297: 195-203. [6] Zeng W, Du J, and Gao W, et al.. Robust moving object segmentation on H.264/AVC compressed video using the block-based MRF model. Real-Time Imaging, 2005, 11(4): 290-299. [7] Mezaris V, Kompatsiaris I, and Boulgouris N V, et al.. Real-time compressed-domain spatiotemporal segmentation and ontologies for video indexing and retrieval[J].IEEE Trans. on Circuits and Systems for Video Technology.2004, 14(5):606-621 [8] Babu R V, Ramakrishnan K R, and Srinivasan S H. Video object segmentation: a compressed domain approach[J].IEEE Trans. on Circuits and Systems for Video Technology.2004, 14(4):462-474 [9] How-Lung E, and Kai-Kuang M. Spatiotemporal segmentation of moving video objects over MPEG compressed domain. IEEE International Conference on Multimedia and Expo, New York, 2000, 3: 1531-1534. [10] Benzougar A, and Bouthemy P, Fablet R. MRF-based moving object detection from MPEG coded video. IEEE International Conference on Image Processing, Thessaloniki, Greece, 2001, 3: 402-405. [11] Babu R V, and Ramakrishnan K R. Compressed domain motion segmentation for video object extraction. IEEE International Conference on Acoustics, Speech, and Signal Processing, Florida, USA, 2002, 4: 3788-3791. [12] Sukmarg O, and Rao K R. Fast object detection and segmentation in MPEG compressed domain. IEEE Region 10 Technical Conference, Malaysia, 2000, 3: 364-368. [13] Comaniciu D and Meer P. Mean shift analysis and applications. IEEE International Conference on Computer Vision, Kerkyra, Greece, 1999, 2: 1197-1203. [14] Christoudias C M, Georgescu B, and Meer P. Synergism in low level vision. International Conference on Pattern Recognition, Quebec City, Canada, 2002, 4: 150-155. [15] Chang H S, and Kang K. A compressed domain scheme for classifying block edge patterns[J].IEEE Trans. on Image Processing.2005, 14(2):145-151 [16] Yeo B L, and Liu B. Rapid scene analysis on compressed video[J].IEEE Trans. on Circuits and Systems for Video Technology.1995, 5(6):533-544 [17] Comaniciu D, and Meer P. Mean shift: a robust approach toward feature space analysis[J].IEEE Trans. on Pattern Analysis and Machine Intelligence.2002, 24(5):1-18 [18] Black M J, and Anandan P. The robust estimation of multiple motions: parametric and piecewise-smooth flow fields[J].Computer Vision and Image Understanding.1996, 63(1):75-104 [19] Li S Z. Markov Random Field Modeling in Image Analysis. Tokyo: Springer-Verlag, 2001: 235-237.

施引文献

资源附件(0)

访问统计