Segmentation of Human Body Parts in Video Frames Based on Intrinsic Distance

  • Yu-Chun Lai
  • Hong-Yuan Mark Liao
  • Cheng-Chung Lin
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4810)

Abstract

We propose an intrinsic-distance based segmentation approach for segmenting human body parts in video frames. First, since the human body can be seen as a set of articulated parts, we utilize the moving articulated attributes to identify body part candidate regions automatically. The candidate regions and the background candidate regions are generated by voting and assigned to the spatiotemporal volume, which is comprised of frames of the video. Then, the intrinsic distance is used to estimate the boundaries of each body part. Our intrinsic distance-based segmentation technique is applied in the spatiotemporal volume to extract the optimal boundaries of the intrinsic distance in a video and obtain segmented frames from the segmented volume. The segmented results show that the proposed approach can tolerate incomplete and imprecise candidate regions because it provides temporal continuity. Furthermore, it can reduce over growing in the original intrinsic distance-based algorithm, since it can handle ambiguous pixels. We expect that this research can provide an alternative to segmenting a sequence of body parts in a video.

Keywords

Segmentation Human body part Intrinsic distance 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Haritaoglu, I., Harwood, D., Davis, L.: Ghost: A Human Body Part Labeling System Using Silhouettes. In: Proc. International Conference of Pattern Recognition, pp. 77–82 (1998)Google Scholar
  2. 2.
    Hsieh, J.-W., Chen, C.-C., Hsu, Y.-T.: Segmentation of Human Body Parts Using Deformable Triangulation. In: IEEE International Conference on Pattern Recognition, vol. 1, pp. 355–358 (2006)Google Scholar
  3. 3.
    Ramanan, D., Forsyth, D.A., Zisserman, A.: Tracking People by Learning Their Appearance. IEEE Transactions on Pattern Analysis and Machine Intelligence 29(1), 65–81 (2007)CrossRefGoogle Scholar
  4. 4.
    Lai, Y.-C., Liao, H.Y.M.: Human Motion Recognition Using Clay Representation of Trajectories. In: Proc. IEEE International Conference on Intelligent Information Hiding and Multimedia Signal Processing, USA (December 2006)Google Scholar
  5. 5.
    Yatziv, L., Sapiro, G.: Fast image and video colorization using chrominance blending. IEEE Transactions on Image Processing 15(5), 1120–1129 (2006)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Yu-Chun Lai
    • 1
  • Hong-Yuan Mark Liao
    • 1
    • 2
  • Cheng-Chung Lin
    • 1
  1. 1.Department of Computer Science, National Chiao-Tung UniversityTaiwan
  2. 2.Institute of Information Science, Academia SinicaTaiwan

Personalised recommendations