The Journal of Korean Institute of Communications and Information Sciences (한국통신학회논문지)

The Korean Institute of Commucations and Information Sciences (한국통신학회)
권호 표지
DOI QR코드

DOI QR Code

Probabilistic Graph Based Object Category Recognition Using the Context of Object-Action Interaction

물체-행동 컨텍스트를 이용하는 확률 그래프 기반 물체 범주 인식

  • Yoon, Sung-baek (Electrical and Computer Engineering, Sungkyunkwan University) ;
  • Bae, Se-ho (Electrical and Computer Engineering, Sungkyunkwan University) ;
  • Park, Han-je (Electrical and Computer Engineering, Sungkyunkwan University) ;
  • Yi, June-ho (Electornic and Electrical Engineering, Sungkyunkwan University, Information and Communication Engineering, North university of China)
  • Received : 2015.07.23
  • Accepted : 2015.10.13
  • Published : 2015.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

The use of human actions as context for object class recognition is quite effective in enhancing the recognition performance despite the large variation in the appearance of objects. We propose an efficient method that integrates human action information into object class recognition using a Bayesian appraoch based on a simple probabilistic graph model. The experiment shows that by using human actions ac context information we can improve the performance of the object calss recognition from 8% to 28%.

다양한 외형 변화를 가지는 물체의 범주 인식성능을 향상 시키는데 있어서 사람의 행동은 매우 효과적인 컨텍스트 정보이다. 본 연구에서는 Bayesian 접근법을 기반으로 하는 간단한 확률 그래프 모델을 통해 사람의 행동을 물체 범주 인식을 위한 컨텍스트 정보로 활용하였다. 다양한 외형의 컵, 전화기, 가위 그리고 스프레이 물체에 대해 실험을 수행한 결과 물체의 용도에 대한 사람의 행동을 인식함으로써 물체 인식 성능을 8%~28%개선할 수 있었다.

Keywords

References

  1. V. Singh and R. Nevatia, "Multiple pose context trees for estimating human pose in object context," in CVPR Workshop on Structural Models in Computer Vision, pp. 17-24, San Francisco, CA, USA, Jun. 2010.
  2. B. Yao and L. FeiFei, "Recognizing human-object interactions in still images by modeling the mutual context of objects and human poses," IEEE Trans. Pattern Anal. Mach. Intell., vol. 34, no. 9, pp. 1691-1703, Sept. 2012. https://doi.org/10.1109/TPAMI.2012.67
  3. D. J. Moore, I. A. Essa, and M. H. Hayes, "Exploiting human actions and object context for recognition tasks," in Proc. Int. Conf. Computer Vision, vol. 1, pp. 80-86, Corfu, Greece, Sept. 1999.
  4. A. Gupta, A. Kembhavi, and L. S. Davis, "Observing human-object interactions: Using spatial and functional compatibility for recognition," IEEE Trans. Pattern Anal. Mach. Intell., vol. 31, no. 10, pp. 1775-1789, Oct. 2009. https://doi.org/10.1109/TPAMI.2009.83
  5. W. Fan, E. Greengrass, and J. McCloskey, "Effective estimation of posterior probabilities: Explaining the accuracy of randomized decision tree approaches," in Proc. Int. Conf. Data Mining, pp. 154-161, Houston, Texas, USA, Nov. 2005.
  6. J. Zhu, H. Zou, S. Rosset, and T. Hastie, "Multi-class Adaboost," Stats. Its Interface, vol. 2, no. 3, pp. 349-360, 2009. https://doi.org/10.4310/SII.2009.v2.n3.a8
  7. L. D. Bourdev and J. Malik, "Poselets: Body part detectors trained using 3d human pose annotations," in Proc. Int. Conf. Computer Vision, pp. 1365-1372, Kyoto, Japan, Sept. 2009.
  8. N. Dalal and B. Triggs, "Histogram of oriented gradients for human detection," in Proc. Computer Vision and Pattern Recognition, pp. 886-893, San Diego, CA, USA, Jun. 2005.
  9. B. W. Chung, K. Y. Park, and S. Y. Hwang, "A fast and efficient Haar-like feature selection algorithm for object detection," J. KICS, vol. 38A, no. 6, pp. 486-491, Jun. 2013. https://doi.org/10.7840/kics.2013.38A.6.486
  10. K. Y. Park and S. Y. Hwang, "An improved normalization method for Haar-like features for real-time object detection," J. KICS, vol. 36, no. 8, pp. 505-515, Aug. 2013.
  11. H. G. Chung and E. S. Kim, "Improved recognition of far objects by using DPM method in curving-effective integral imaging," J. KICS, vol. 37A, no. 2, pp. 128-134, Feb. 2013.

Cited by

  1. 다중 디지털 신호의 비교를 위한 병렬 기법의 VLSI 설계 vol.21, pp.4, 2015, https://doi.org/10.6109/jkiice.2017.21.4.781
  2. Topological and Statistical Analysis for the High-Voltage Transmission Networks in the Korean Power Grid vol.42, pp.4, 2017, https://doi.org/10.7840/kics.2017.42.4.923