Presentation 2011/9/14
A Study on Rhythm Extraction from Dance Video and Its Evaluation based on Principal Component Analysis
Tomoaki Konno, Jianfeng Xu, Koichi Takagi, Shigeyuki Sakazawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a rhythm extraction method utilizing short term principal component analysis (ST-PCA) for motion video in order to automatically evaluate rhythm of a dance. Especially, rhythm features are extracted without explicit background separation by ST-PCA for each segment of entire video sequences, which consists of cyclic motion of basic dances. Furthermore, we enhance the method for online-evaluation of the dance rhythm by comparing the rhythm between a student and a teacher from segment to segment. Our experimental results have shown that our method was effective for online rhythm evaluation with latency less than 2 seconds. The accuracy between visual evaluation and objective one was 0.86, when ST-PCA with analysis windows with twice length of the cycle was applied.
Keyword(in Japanese) (See Japanese page)
Keyword(in English)
Paper #
Date of Issue

Conference Information
Committee SIS
Conference Date 2011/9/14(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Smart Info-Media Systems (SIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Study on Rhythm Extraction from Dance Video and Its Evaluation based on Principal Component Analysis
Sub Title (in English)
Keyword(1)
1st Author's Name Tomoaki Konno
1st Author's Affiliation KDDI R&D Laboratories Inc.()
2nd Author's Name Jianfeng Xu
2nd Author's Affiliation KDDI R&D Laboratories Inc.
3rd Author's Name Koichi Takagi
3rd Author's Affiliation KDDI R&D Laboratories Inc.
4th Author's Name Shigeyuki Sakazawa
4th Author's Affiliation KDDI R&D Laboratories Inc.
Date 2011/9/14
Paper #
Volume (vol) vol.111
Number (no) 210
Page pp.pp.-
#Pages 6
Date of Issue