Detecting Human Actions by 3D Deformable Parts Models

Ming Yang; Yong Yang

doi:10.4028/www.scientific.net/AMR.981.331

Paper Titles

A Clock Optimization Method in the Digital Zooming of the Image Signal Processing System
p.315

A Novel Rapid Full-View Measuring Method by Coded Structured Light
p.319

Design and Implementation of Unified DCT/IDCT Architecture Based on FPGA
p.323

Diffusion-Substitution Mechanism for Color Image Encryption Based on Multiple Chaotic Systems
p.327

Detecting Human Actions by 3D Deformable Parts Models
p.331

Video Segmentation Algorithm Based on Improved Kirsch Edge Operator and Three-Frame Difference
p.335

Denoise PET Images Based on a Combining Method of EMD and ICA
p.340

An Improved C-Means Clustering for Image Segmentation
p.344

Coded-Structured Light System Calibration Using Orthogonal Phase Shift Coding Combined with Zhang’s Method
p.348

HomeAdvanced Materials ResearchAdvanced Materials Research Vol. 981Detecting Human Actions by 3D Deformable Parts...

Detecting Human Actions by 3D Deformable Parts Models

Abstract:

In this paper, we introduce the high performance Deformable part models from object detection into human action recognition and localization and propose a unified method to detect action in video sequences. The Deformable part models have attracted intensive attention in the field of object detection. We generalize the approach from 2D still images to 3D spatiotemporal volumes. The human actions are described by 3D histograms of oriented gradients based features. Different poses are presented by mixture of models on different resolutions. The model autonomously selects the most discriminative 3D parts and learns their anchor positions related to the root. Empirical results on several video datasets prove the efficacy of our proposed method on both action recognition and localization.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Advanced Materials Research (Volume 981)

Pages:

331-334

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.981.331

Citation:

Cite this paper

Online since:

July 2014

Authors:

Ming Yang*, Yong Yang

Keywords:

3D HOG, Action Recognition, Deformable Parts, Latent SVM

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] Laptev I.: On Space-time Interest Points, International Journal of Computer Vision, pp.107-123, (2005).

Google Scholar

[2] Lowe D.G.: Distinctive Image Features from Scale-invariant Keypoints, International Journal of Computer Vision, pp.91-110, (2004).

DOI: 10.1023/b:visi.0000029664.99615.94

Google Scholar

[3] Yuelei X., Hong C., Zhe L., Luhong L., Xilin C. and Debin Z.: A Unified Framework for Locating and Recognizing Human Actions, CVPR, pp.25-32, (2011).

Google Scholar

[4] Felzenszwalb P.F., Girshick R.B., McAllester D. and Ramanan D.: Object Detection with Discriminatively Trained Part-based Models, IEEE Trans. on PAMI, pp.1627-1645, (2010).

DOI: 10.1109/tpami.2009.167

Google Scholar

[5] M. Everingham, L. Van Gool, C. K. Williams, J. Winn, and A. Zisserman: The PASCAL Visual Object Classes Challenge 2007 (VOC2007) Results, http: /www. pascal- network. org/challenges/VOC/voc2007.

DOI: 10.1007/s11263-009-0275-4

Google Scholar

[6] Yang Wang, Mori G.: Hidden Part Models for Human Action Recognition: Probabilistic versus Max Margin, IEEE Trans. on PAMI, pp.1310-1323, (2011).

DOI: 10.1109/tpami.2010.214

Google Scholar

[7] Klaser, Alexander, and Marcin Marszalek.: A spatio-temporal descriptor based on 3D-gradients, British Machine Vision Conference, pp.995-1004, (2008).

DOI: 10.5244/c.22.99

Google Scholar