Advertisement

GPU-Based Approach for Human Action Recognition in Video

  • Ishita DuttaEmail author
  • Vikas Tripathi
  • Vaishali Dabral
  • Pooja Sharma
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 713)

Abstract

The power of graphic processing units (GPUs) can be harnessed to obtain an appreciable increase in computing performances by parallelizing various techniques. In view of this, the paper compares the performance of various descriptive statistical techniques like mean, variance and standard deviation on GPU for centroid calculation. Taking after this, the most productive procedure from the said methods has been contrasted with centroid calculation using k-means, processed on CPU. An appreciable increase in accuracy was achieved when we processed the above-mentioned techniques on GPU for centroid calculation in comparison with centroid calculation processed on CPU using k-means technique. The HMDB-51 dataset is used for computations. The aim of the paper is to find the most efficient and accurate approach for centroid and distance calculation in clustering. We attain an accuracy enhancement of 6.58% on comparing centroid calculation using variance method on GPU to centroid calculation using k-means on CPU.

Keywords

GPU Statistical techniques HMDB-51 

References

  1. 1.
    Moeslund, T.: Summaries of 107 computer vision-based human motion capture papers Technical Report LIA 99-01, University of Aalborg (1999)Google Scholar
  2. 2.
    Pang, X., Wang, L., Wang, X., Qiao, Y.: Bag of visual words and fusion methods for action recognition; comprehensive study and good practice. Comput. Vis. Image Underst. 109–125 (2016)Google Scholar
  3. 3.
    Ganti, V., Gehrke, J., Ramakrishnan, R.: Mining very large databases. Computer 32, 38–45 (1999)CrossRefGoogle Scholar
  4. 4.
    Judd, D., McKinley, P., Jain, A.: Large-scale parallel data clustering. In: Proceedings of the International Conference on Pattern Recognition, pp. 488–493 (1996)Google Scholar
  5. 5.
    Yadav, K., Mittal, A., Ansari, M.A., Vishwarup, V.: Parallel implementation of similarity measures on GPU architecture using CUDA. Indian J. Comput. Sci. Eng. (IJCSE)Google Scholar
  6. 6.
    Dhillon, I.S., Modha, D.S.: A data clustering algorithm on distributed memory multiprocessors. Large-scale parallel data mining. Lect. Notes Artif. Intell. 1759, 245–260 (2000)Google Scholar
  7. 7.
    Goil, S., Nagesh, H., Chaudhary, A.: MAFIA: efficient and scalable subspace clustering for very large datasets. Technical Report No. CPDC-TR-9906-010 (1999)Google Scholar
  8. 8.
    Ester, M., Kriegel, H.P., Sanders, J., Xu, X.: A density based algorithm for discovering clusters in large spatial databases with noise. In: Proceedings of the 2nd International Conference in Knowledge Discovery in Databases and Datamining (1996)Google Scholar
  9. 9.
    Nagesh, H., Goil, S., Choudhary, A.: A scalable parallel subspace clustering algorithm for massive data sets. In: Proceedings International Conference on Parallel Processing, pp. 477–484. IEEE Computer Society (2000)Google Scholar
  10. 10.
    Ng, M.K., Zhexue, H.: A parallel k-prototypes algorithm for clustering large data sets in data mining. Intell. Data Eng. Learn. 3, 263–290 (1999)Google Scholar
  11. 11.
    Kilian, S., Belkoniene, A.: Parallel k/h-means clustering for large data sets. In: Euro-Par’99 Parallel Processing, pp. 1451–1454. Springer, Berlin, Heidelberg (1999)Google Scholar
  12. 12.
    Alsmirat, M.A., Jararweh, Y., Al-Ayyoub, M., Shehab, M.A., Gupta, B.B.: Accelerating compute intensive medical imaging segmentation algorithms using hybrid CPU-GPU implementations. Multimed. Tools Appl. 76, 3537–3555 (2017)CrossRefGoogle Scholar
  13. 13.
    Dabral, V., Tripathi, V., Khan, K.: Real time computation of clustering and distance matrix through GPU. Int. J. Control Appl. 9, 583–590 (2017)Google Scholar
  14. 14.
    Tripathi, V., Gangodkar, D., Latta, V., Mittal. A.: Robust abnormal event recognition via motion and shape analysis at ATM installations. J. Electr. Comput. Eng. (2015)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  • Ishita Dutta
    • 1
    Email author
  • Vikas Tripathi
    • 2
  • Vaishali Dabral
    • 3
  • Pooja Sharma
    • 2
  1. 1.Indian Institute of Engineering Science and Technology, ShibpurHowrahIndia
  2. 2.Graphic Era UniversityDehradunIndia
  3. 3.Indraprastha Institute of Information Technology, DelhiNew DelhiIndia

Personalised recommendations