[go: up one dir, main page]
More Web Proxy on the site http://driver.im/ Skip to main content
Springer Nature Link
Log in

Human body segmentation based on deformable models and two-scale superpixel

  • Theoretical Advances
  • Published:
Pattern Analysis and Applications Aims and scope Submit manuscript

    We’re sorry, something doesn't seem to be working properly.

    Please try refreshing the page. If that doesn't work, please contact support so we can address the problem.

Abstract

In this paper, we propose a novel method to segment human body in static images by graph cuts based on two deformable models at two-scale superpixel. In our study, body segmentation is decomposed into torso detection and lower body recovery. Based on the first-scale superpixel, the seeds of torso are obtained on the basis of the coarse torso region, which is estimated by an improved deformable torso model. For the lower body, we estimate the hip region to obtain the seeds of lower body at the second-scale superpixel. Besides, a deformable upper leg model is designed to derive more foreground seeds of the lower body. To avoid failure caused by the heavy dependence between the two hierarchies, a scheme of probabilistic hierarchical detection is presented. Experiments on our datasets containing 200 images photographed by ourselves and 100 other images collected from public datasets show that our approach can accurately segment human body in static images with a variety of poses, backgrounds and clothing. Segmenting the human body in static image based on deformable torso and upper leg models at two-scale superpixel.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
£29.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price includes VAT (United Kingdom)

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13

Similar content being viewed by others

Notes

  1. In our experiments, it is set to the half of the covariance of the χ 2 distances \( d_{{\chi^{2} }} (S_{i}^{t} ,T_{s} ) \).

  2. When the estimation of the likelihood is performed, the bounding box of torso is fixed ignoring the face location, as well as the GMM of the background.

  3. The ground truth of images derived from Berkeley and INRIA are labeled by hand. We select the images according to the face. That is, if the face in the image can be detected, we then choose this image.

  4. We download the code from http://www.seas.upenn.edu/~timothee/software/ncut_multiscale/ncut_multiscale.html

  5. To the best of our knowledge, there is no evaluation criterion for this. We simply count the detected limbs for comparison.

References

  1. Boykov Y, Veksler O, Zabih R (2001) Fast approximate energy minimisation via graph cuts. IEEE Trans PAMI 23(11):1222–1239

    Article  Google Scholar 

  2. Shi J, Malik J (2000) Normalized cuts and image segmentation. IEEE Trans PAMI 22(8):888–905

    Article  Google Scholar 

  3. Mori G, Malik J (2002) Estimating human body configurations using shape context matching. In: European conference on computer vision. LNCS 2352, 666–680

  4. Poppe R, Poel M (2005) Example-based pose estimation in monocular images using compact fourier descriptors. Technical report TR-CTIT-05-49, pp 1–19

  5. Agarwal A, Triggs B (2006) A local basis representation for estimating human pose from cluttered images. ACCV, pp 50–59

  6. Lin Z, Davis LS, Doermann D, DeMenthon D (2007) Hierarchical Part-Template Matching for Human Detection and Segmentation, ICCV

  7. Mohan A, Papageorgiou C, Poggio T (2001) Example-based object detection in images by components. IEEE Trans PAMI 23(4):349–361

    Article  Google Scholar 

  8. Mori G, Ren X, Efros AA, Malik J (2004) Recovering human body configurations: combining segmentation and recognition. CVPR, pp 326–333

  9. Hu Z, Lin X, Yan H (2006) Torso Detection in Static Images. ICSP

  10. Strinivasan P, Shi J (2007) Bottom-up recognition and parsing of the human body. CVPR

  11. Ramanan D, Forsyth DA (2003) Finding and tracking people from the bottom up. CVPR

  12. Lee MW, Cohen I (2004) Human upper body pose estimation in static images. ECCV

  13. Ramanan D, Sminchisescu C (2006) Training deformable models for localization. CVPR, 206–213

  14. Mori G (2005) Guiding model search using segmentation. ICCV, pp 1417–1423

  15. Lee MW, Nevatia R (2007) Body part detection for human pose estimation and tracking. IEEE WMVC, pp 23–31

  16. Hu Z, Wang G, Lin X, yan H (2009) Recovery of upper body poses in static images based on joints detection. Pattern Recog Lett 30(5):503–512

    Article  Google Scholar 

  17. Fulkerson B, Vedaldi A, Soatto S (2009) Class segmentation and object localization with superpixel neighborhoods. ICCV, pp 670–678

  18. Ren X, Malik J (2003) Learning a classification model for segmentation. ICCV, pp 10–17

  19. Oliveira PR, Romero RAF (2008) Improvements on ICA mixture models for image pre-processing and segmentation. Neurocomputing 71(10-12):2180–2193

    Article  Google Scholar 

  20. Maire M, Arbelaez P, Fowlkes C, Malik J (2008) Using contours to detect and localize junctions in natural images. CVPR, pp 1–8

  21. Viola P, Jones M (2004) Robust real-time face detection. Int J Comput Vis 57(2):137–154

    Article  Google Scholar 

  22. Rother C, Kolmogorov V, Blake A (2004) Grabcut: interactive foreground extraction using iterated graph cuts. ACM Trans Graph 23(3):309–314

    Article  Google Scholar 

Download references

Acknowledgment

The work was supported by the Fundamental Research Funds for the Central Universities, No. DUT10JS05, the National Natural Science Foundation of China (NSFC), No. 61071209 and Omron Corporation's grant No. DUT09001.

Author information

Authors and Affiliations

  1. Department of Electronic Engineering, Dalian University of Technology, Dalian, 116023, China

    Shifeng Li & Hu-Chuan Lu

  2. Digtial Imaging and Mobile Devices Department Image Processing Design Section, Omron, Kyoto, Japan

    Xiang Ruan

  3. College of Information Science and Engineering, Ritsumeikan University, Kusatsu, 525-8577, Japan

    Yen-Wei Chen

Authors
  1. Shifeng Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Hu-Chuan Lu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Xiang Ruan
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yen-Wei Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Hu-Chuan Lu.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (DOC 853 kb)

Rights and permissions

Reprints and permissions

About this article

Cite this article

Li, S., Lu, HC., Ruan, X. et al. Human body segmentation based on deformable models and two-scale superpixel. Pattern Anal Applic 15, 399–413 (2012). https://doi.org/10.1007/s10044-011-0220-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10044-011-0220-3

Keywords

Search

Navigation