Kinect Depth Data Segmentation Based on Gauss Mixture Model Clustering

Abstract:

Indoor scene understanding based on the depth image data is a cutting-edge issue in the field of three-dimensional computer vision. Taking the layout characteristics of the indoor scenes and more plane features in these scenes into account, this paper presents a depth image segmentation method based on Gauss Mixture Model clustering. First, transform the Kinect depth image data into point cloud which is in the form of discrete three-dimensional point data, and denoise and down-sample the point cloud data; second, calculate the point normal of all points in the entire point cloud, then cluster the entire normal using Gaussian Mixture Model, and finally implement the entire point clouds segmentation by RANSAC algorithm. Experimental results show that the divided regions have obvious boundaries and segmentation quality is above normal, and lay a good foundation for object recognition.

• Periodical: Advanced Materials Research (Volumes 760-762)
• Pages: 1556-1561
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.760-762.1556
• Online since: September 2013
• Authors: Ting Wei Du, Bo Liu
• Keywords: Depth Data Segmentation, Gauss Mixture Model, Indoor Scene Understanding, KINECT, RANSAC Algorithm

References

[1] S. Guðmundsson, J. Sveinsson, M. Pardas, Model-Based Hand Gesture Tracking in ToF Image Sequences,. Perales and R.B. Fisher (Eds. ): AMDO 2010, LNCS 6169, p.118–127, (2010).

DOI: 10.1007/978-3-642-14061-7_12

[2] S. Oprisescu, C. Burlacu, V. Buzuloiu, Action Recognition using Time of Flight Cameras, 8th Int. COMM, (2010).

DOI: 10.1109/iccomm.2010.5509074

[3] J. Civit, O. D. Escoda, Robust Foreground Segmentation for GPU Architecture in an Immersive 3D Video-Conferencing Systems, IEEE Int. Workshop on MMSP, (2010).

DOI: 10.1109/mmsp.2010.5661997

[4] Jain, A. K., Topchy, A., Law, M. H. C., and Buhmann, Landscape of clus-tering algorithms. In International Conference on Pattern Recognition (ICPR 2004), p.260–263. (2004).

DOI: 10.1109/icpr.2004.1334073

[5] Brice, C.R. and Fennema,C. L. Scene analysis using regions. Artificial Intelli-gence, 1(3-4): 205–226. (1970).

DOI: 10.1016/0004-3702(70)90008-1

[6] Pavlidis, T. Structural Pattern Recognition. Springer-Verlag, Berlin; New York. (1977).

[7] Cremers, D., Rousson, M., and Deriche, R. A review of statistical approaches to level set segmentation: integrating color, texture, motion and shape. International Journal of Computer Vision, 72(2): 195–215. (2007).

DOI: 10.1007/s11263-006-8711-1

[8] Besl P J, Jain R C. Segmentation Through Variable Order Surface Fitting. IEEE Transactions on Pattern Analysis and Machine Intelligence, 10(2): 167-192. (1988).

DOI: 10.1109/34.3881

[9] Hoffman R L, Jain A K. Segmentation and classification of range images. IEEE Transactions on Pattern Analysis and Machine Intelligence, 9( 5): 608~ 620. (1987).

DOI: 10.1109/tpami.1987.4767955

[10] Gotardo P, Bellon O. Range Image Segmentation Into Planar and Quadric Surfaces Using an Improved Robust Estimator and Genetic Algorithm. IEEE Transactions on System, Man, Cybernetics B, 34( 6): 2303~ 2316. (2004).

DOI: 10.1109/tsmcb.2004.835082

[11] Mirante E, Georgiev M, Gotchev A. A fast image segmentation algorithm using color and depth map. 3DTV Conference: The True Vision - Capture, Transmission and Display of 3D Video (3DTV-CON), p.1~4. (2011).

DOI: 10.1109/3dtv.2011.5877227

[12] X.M. Chen L.T. Jiang. Research of 3D reconstruction and filtering algorithm based on depth information of kinect. Application Research of Computers. vol. 29. (2012).

[13] Fraley C, Raftery AE . How many clusters? Which clustering methods? Answers via model-based cluster analysis. Computer Journal, 41, 8, 578~588. (1998).

DOI: 10.1093/comjnl/41.8.578

[14] Fischler M A , Bolles R C . Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography. Communications of the ACM. 24(6): 381–395. (1981).

DOI: 10.1145/358669.358692