Skeleton-augmented human action understanding by learning with progressively refined data

Shih En Wei, Nick C. Tang, Yen-Yu Lin, Ming Fang Weng, Hong Yuan Mark Liao

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

With the aim at accurate action video retrieval, we firstly present an approach that can infer the implicit skeleton structure for a query action, an RGB video, and then propose to expand this query with the inferred skeleton for improving the performance of retrieval. It is inspired by the observation that skeleton structures can compactly and effectively represent human actions, and are helpful in bridging the semantic gap in action retrieval. The focal point is hence on action skeleton estimation in RGB videos. Specifically, an iterative training procedure is developed to select relevant training data for inferring the skeleton of an input action, since corrupt training data not only degrades performance but also complicates the learning process. Through the iterations, relevant training data are gradually revealed, while more accurate skeletons are inferred with the refined training set. The proposed approach is evaluated on ChaLearn 2013. Significant performance gains in action retrieval are achieved with the aid of the inferred skeletons.

Original languageEnglish
Title of host publicationHuEvent 2014 - Proceedings of the 2014 Workshop on Human Centered Event Understanding from Multimedia
PublisherAssociation for Computing Machinery, Inc
Pages7-10
Number of pages4
ISBN (Electronic)9781450331203
DOIs
StatePublished - 7 Nov 2014
Event1st ACM International Workshop on Human Centered Event Understanding from Multimedia, HuEvent 2014 - Orlando, United States
Duration: 7 Nov 2014 → …

Publication series

NameHuEvent 2014 - Proceedings of the 2014 Workshop on Human Centered Event Understanding from Multimedia

Conference

Conference1st ACM International Workshop on Human Centered Event Understanding from Multimedia, HuEvent 2014
CountryUnited States
CityOrlando
Period7/11/14 → …

Keywords

  • Action retrieval
  • Pose estimation

Fingerprint Dive into the research topics of 'Skeleton-augmented human action understanding by learning with progressively refined data'. Together they form a unique fingerprint.

Cite this