An Efficient Method of Key-Frame Extraction Based on a Cluster Algorithm

• Authors: Qiang Zhang , Shao-Pei Yu , Dong-Sheng Zhou , Xiao-Peng Wei
• Languages of publication: EN

Abstracts

EN

This paper proposes a novel method of key-frame extraction for use with motion capture data. This method is based on an unsupervised cluster algorithm. First, the motion sequence is clustered into two classes by the similarity distance of the adjacent frames so that the thresholds needed in the next step can be determined adaptively. Second, a dynamic cluster algorithm called ISODATA is used to cluster all the frames and the frames nearest to the center of each class are automatically extracted as key-frames of the sequence. Unlike many other clustering techniques, the present improved cluster algorithm can automatically address different motion types without any need for specified parameters from users. The proposed method is capable of summarizing motion capture data reliably and efficiently. The present work also provides a meaningful comparison between the results of the proposed key-frame extraction technique and other previous methods. These results are evaluated in terms of metrics that measure reconstructed motion and the mean absolute error value, which are derived from the reconstructed data and the original data.

Keywords

EN

Motion Capture; ISODATA; Adaptive Threshold

• Publisher: De Gruyter Open
• Journal: Journal of Human Kinetics
• Year: 2013
• Volume: 39
• Issue: 1
• Pages: 5-14

Dates

published

1 - 12 - 2013

online

31 - 12 - 2013

Contributors

author

Qiang Zhang

• Key Laboratory of Advanced Design and Intelligent Computing (Dalian University), Ministry of Education, Dalian, China.

author

Shao-Pei Yu

• Key Laboratory of Advanced Design and Intelligent Computing (Dalian University), Ministry of Education, Dalian, China.

author

Dong-Sheng Zhou

• Key Laboratory of Advanced Design and Intelligent Computing (Dalian University), Ministry of Education, Dalian, China.

author

Xiao-Peng Wei

• Key Laboratory of Advanced Design and Intelligent Computing (Dalian University), Ministry of Education, Dalian, China.

References

• Baak A, Müeller M, Seidel HP. An efficient algorithm for keyframe-based motion retrieval in the presence of temporal deformations. New York: 1st ACM international Conference on Multimedia Information Retrieval, 451-458; 2008
• Bulut E, Capin T. Keyframe extraction from motion capture data by curve saliency. Belgium: Proceedings of 20th Annual Conference on Comput Animat and Social Agents, 63-67; 2007
• Clifford KFS, Baciu G. Entropy-based Motion Extraction for Motion Capture Animation: Motion Capture and Retrieval. Comput Animat Virt W, 2005; 16(3-4): 225-235
• CMU Graphics Lab Motion Capture Database, 2013. Available at: http://mocap.cs.cmu.edu; accessed on 09.15.2013
• Cooper M, Foote J. Summarizing video using nonnegative similarity matrix factorization. US Virgin Islands:IEEE Workshop on Multimedia Signal Processing, 25-28; 2002
• Gong YH, Liu X. Video summarization using singular value decomposition. South Carolina: Proceedings IEEE Conference on Computer Vision and Pattern Recognition, 174-180; 2000
• Halit C, Capin T. Multiscale Motion Saliency for Keyframe Extraction from Motion Capture Sequences. Comput Animat Virt W, 2011; 22(1): 3-14[WoS][Crossref]
• Huang KS, Chang CF, Hsu YY, Yang SN. Key Probe: a technique for animation keyframe extraction. The Visual Comput, 2005; 21(8-10): 532-541
• Li S, Okuda M, Takahashi S. Embedded keyframe extraction for CG animation by frame decimation. Amsterdam: Proceedings of IEEE Int. Conference on Multimedia & Expo, 1404-1407; 2005
• Lim IS, Thalmann D. Key-Posture Extraction out of Human Motion Data by Curve Simplification. Istanbul:23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1167-1169; 2001
• Liu F, Zhuang YT, Wu F, Pan YH. 3D Motion Retrieval with Motion Index Tree. Comput Vis Image Und, 2003; 92(2-3): 265-284[Crossref]
• Lowe DG. Three-dimensional object recognition from single two dimensional images. Artifl Intell, 1987; 31(3): 355-395
• Memarsadeghi N, Mount DM, Netanyahu NS, Le MJ. A fast implementation of the ISODATA clustering algorithm. Int J Comput Geom Ap, 2007; 17(1): 71-103[Crossref]
• Morishima S, Kuriyama S, Kawamoto S, Suzuki T, Taira M, Yotsukura T, Nakamura S. Data-driven efficient production of cartoon character animation. New York: ACM SIGGRAPH Sketches, 119; 2007
• Park MJ, Shin SY. Example-based motion cloning. Comput Animat Virt W, 2004; 15(3-4): 245-257[Crossref]
• Phillips RD, Watson LT, Wynne RH. A study of fuzzy clustering within the IGSCR framework. Auburn: Proceedings of the 46th Annual Southeast Regional Conference, 140-145; 2008
• Takaki M, Tamura K, Mori Y, Kitakami H. A Extraction Method of Overlapping Cluster based on Network Structure Analysis. Silicon Valley: IEEE/WIC/ACM International Conferences on Web Intelligent Agent Technology, 212-216; 2007
• Tilmanne J, Hidot S, Ravet T. Mockey: motion capture as a tool for keyframing animation. QPSR of the numediart research program, 2009; 2(4): 119-124
• Togawa H, Okuda M. Position-based keyframe selection for human motion animation. Fukuoka: Proceedings of the 11th International Conference on Parallel and Distributed Systems, 182-185; 2005
• Wu L, Li X, Yong S. A New Method for Bad Data Identification of Integrated Power System in Warship Based on Fuzzy ISODATA Clustering Analysis. Electr Eng, 2011; 97: 101-108
• Xiao J, Zhuang YT, Yang T, Wu F．An Efficient Keyframe Extraction from Motion Capture Data. Berlin: Springer Heidelberg Advances in Computer Graphics, 494-501; 2006

Identifiers

DOI

10.2478/hukin-2013-0063