Recognition of Pansori Motion in Still Images Based on Keypoint Detection 


Vol. 47,  No. 4, pp. 575-583, Apr.  2022
10.7840/kics.2022.47.4.575


PDF
  Abstract

Artificial intelligence research on music area in MIR(Music Information Retrieval) is being conducted very actively. This paper aims to recognize the Balim motions, an important component of Pansori, and proposes a method to automatically classify them from Pansori video. In the proposed method, the region of a singer is at first detected from every video frame, then the Balim motion is classified based on still image analysis in the detected region. For detecting singer object we use the transformer-based object detection, which has an excellent performance with high speed for the uniform size of small number of objects, and then try to classify Balim motion with skeletal information of key points including the singer’s joint and fan position. Due to insufficient Pansori Balim motion data, transfer learning using MS COCO data and fine-tuning are performed. The accuracy of 84.8% has been achieved in the classification of Balim motion largely categorized into 7 classes.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

W. Wu, H. Lee, J. Lee, "Recognition of Pansori Motion in Still Images Based on Keypoint Detection," The Journal of Korean Institute of Communications and Information Sciences, vol. 47, no. 4, pp. 575-583, 2022. DOI: 10.7840/kics.2022.47.4.575.

[ACM Style]

Wenqin Wu, Hyejeong Lee, and Joonwhoan Lee. 2022. Recognition of Pansori Motion in Still Images Based on Keypoint Detection. The Journal of Korean Institute of Communications and Information Sciences, 47, 4, (2022), 575-583. DOI: 10.7840/kics.2022.47.4.575.

[KICS Style]

Wenqin Wu, Hyejeong Lee, Joonwhoan Lee, "Recognition of Pansori Motion in Still Images Based on Keypoint Detection," The Journal of Korean Institute of Communications and Information Sciences, vol. 47, no. 4, pp. 575-583, 4. 2022. (https://doi.org/10.7840/kics.2022.47.4.575)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880