Cover Image

Who Danced Better? Ranked TikTok Dance Video Dataset and Pairwise Action Quality Assessment Method

Video-based action quality assessment (AQA) is a non-trivial task due to the subtle visual differences between data produced by experts and non-experts. Current methods are extended from the action recognition domain, where most are based on temporal pattern matching. AQA has additional requirements...

Full description

Saved in:
Bibliographic Details
Main Authors: Irwandi, Hipiny, Hamimah, Ujir, Aidil Azli, Alias, Musdi, Shanat, Mohamad Khairi, Ishak
Format: Article
Language:English
Published: Universitas Ahmad Dahlan 2023
Subjects:
QA75 Electronic computers. Computer science
Online Access:http://ir.unimas.my/id/eprint/41536/2/Who%20Danced%20-%20Copy.pdf
http://ir.unimas.my/id/eprint/41536/
https://ijain.org/index.php/IJAIN/article/view/919
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.unimas.ir.41536
record_format eprints
spelling my.unimas.ir.415362023-08-02T02:51:32Z http://ir.unimas.my/id/eprint/41536/ Who Danced Better? Ranked TikTok Dance Video Dataset and Pairwise Action Quality Assessment Method Irwandi, Hipiny Hamimah, Ujir Aidil Azli, Alias Musdi, Shanat Mohamad Khairi, Ishak QA75 Electronic computers. Computer science Video-based action quality assessment (AQA) is a non-trivial task due to the subtle visual differences between data produced by experts and non-experts. Current methods are extended from the action recognition domain, where most are based on temporal pattern matching. AQA has additional requirements where order and tempo matter for rating the quality of an action. We present a novel dataset of ranked TikTok dance videos and a pairwise AQA method for predicting which video of a same-label pair was sourced from the better dancer. Exhaustive pairings of same-label videos were randomly assigned to 100 human annotators, ultimately producing a ranked list per label category. Our method relies on a successful detection of the subject’s 2D pose inside successive query frames where the order and tempo of actions are encoded inside a produced String sequence. The detected 2D pose returns a top-matching Visual word from a Codebook to represent the current frame. Given a same-label pair, we generate a String value of concatenated Visual words for each video. By computing the edit distance score between each String value and the Gold Standard’s (i.e., the top-ranked video(s) for that label category), we declare the video with the lower score as the winner. The pairwise AQA method is implemented using two schemes, i.e., with and without text compression. Although the average precision for both schemes over 12 label categories is low, at 0.45 with text compression and 0.48 without, precision values for several label categories are comparable to past methods (median: 0.47, max: 0.66). Universitas Ahmad Dahlan 2023-03 Article PeerReviewed text en http://ir.unimas.my/id/eprint/41536/2/Who%20Danced%20-%20Copy.pdf Irwandi, Hipiny and Hamimah, Ujir and Aidil Azli, Alias and Musdi, Shanat and Mohamad Khairi, Ishak (2023) Who Danced Better? Ranked TikTok Dance Video Dataset and Pairwise Action Quality Assessment Method. International Journal of Advances in Intelligent Informatics, 9 (1). pp. 1-12. ISSN 2548-3161 https://ijain.org/index.php/IJAIN/article/view/919
institution Universiti Malaysia Sarawak
building Centre for Academic Information Services (CAIS)
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Malaysia Sarawak
content_source UNIMAS Institutional Repository
url_provider http://ir.unimas.my/
language English
topic QA75 Electronic computers. Computer science
spellingShingle QA75 Electronic computers. Computer science
Irwandi, Hipiny
Hamimah, Ujir
Aidil Azli, Alias
Musdi, Shanat
Mohamad Khairi, Ishak
Who Danced Better? Ranked TikTok Dance Video Dataset and Pairwise Action Quality Assessment Method
description Video-based action quality assessment (AQA) is a non-trivial task due to the subtle visual differences between data produced by experts and non-experts. Current methods are extended from the action recognition domain, where most are based on temporal pattern matching. AQA has additional requirements where order and tempo matter for rating the quality of an action. We present a novel dataset of ranked TikTok dance videos and a pairwise AQA method for predicting which video of a same-label pair was sourced from the better dancer. Exhaustive pairings of same-label videos were randomly assigned to 100 human annotators, ultimately producing a ranked list per label category. Our method relies on a successful detection of the subject’s 2D pose inside successive query frames where the order and tempo of actions are encoded inside a produced String sequence. The detected 2D pose returns a top-matching Visual word from a Codebook to represent the current frame. Given a same-label pair, we generate a String value of concatenated Visual words for each video. By computing the edit distance score between each String value and the Gold Standard’s (i.e., the top-ranked video(s) for that label category), we declare the video with the lower score as the winner. The pairwise AQA method is implemented using two schemes, i.e., with and without text compression. Although the average precision for both schemes over 12 label categories is low, at 0.45 with text compression and 0.48 without, precision values for several label categories are comparable to past methods (median: 0.47, max: 0.66).
format Article
author Irwandi, Hipiny
Hamimah, Ujir
Aidil Azli, Alias
Musdi, Shanat
Mohamad Khairi, Ishak
author_facet Irwandi, Hipiny
Hamimah, Ujir
Aidil Azli, Alias
Musdi, Shanat
Mohamad Khairi, Ishak
author_sort Irwandi, Hipiny
title Who Danced Better? Ranked TikTok Dance Video Dataset and Pairwise Action Quality Assessment Method
title_short Who Danced Better? Ranked TikTok Dance Video Dataset and Pairwise Action Quality Assessment Method
title_full Who Danced Better? Ranked TikTok Dance Video Dataset and Pairwise Action Quality Assessment Method
title_fullStr Who Danced Better? Ranked TikTok Dance Video Dataset and Pairwise Action Quality Assessment Method
title_full_unstemmed Who Danced Better? Ranked TikTok Dance Video Dataset and Pairwise Action Quality Assessment Method
title_sort who danced better? ranked tiktok dance video dataset and pairwise action quality assessment method
publisher Universitas Ahmad Dahlan
publishDate 2023
url http://ir.unimas.my/id/eprint/41536/2/Who%20Danced%20-%20Copy.pdf
http://ir.unimas.my/id/eprint/41536/
https://ijain.org/index.php/IJAIN/article/view/919
_version_ 1773547909317394432
score 13.211869

Similar Items