Modeling and Recognizing Action Contexts in Persons Using Sparse Representation

Kai Ting Chuang*, Jun-Wei Hsieh, Yilin Yan

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

4 Scopus citations


This paper proposes a novel dynamic sparse representation-based classification scheme to treat the problem of interaction action analysis between persons using sparse representation. The occlusion problem and the difficulty to model complicated interactions are the major challenges in person-to-person action analysis. To address the occlusion problem, the proposed scheme represents an action sample in an over-complete dictionary whose base elements are the training samples themselves. This representation is naturally sparse and makes errors (caused by different environmental changes like lighting or occlusions) sparsely appear in the training library. Because of the sparsity, it is robust to occlusions and lighting changes. The difficulty of complicated action modeling can be tackled by adding more examples to the over-complete dictionary. Thus, even though the interaction relations are complicated, the proposed method still works successfully to recognize them and can be easily extended to analyze action events among multiple persons.

Original languageEnglish
Title of host publicationAdvances in Intelligent Systems and Applications - Volume 2
Subtitle of host publicationProceedings of the International Computer
EditorsChang Ruay-Shiung, Peng Sheng-Lung, Lin Chia-Chen
Number of pages11
StatePublished - 28 Jun 2013

Publication series

NameSmart Innovation, Systems and Technologies
ISSN (Print)2190-3018
ISSN (Electronic)2190-3026


  • Daily events
  • Occlusions
  • Sparse Coding
  • Sparse representation

Fingerprint Dive into the research topics of 'Modeling and Recognizing Action Contexts in Persons Using Sparse Representation'. Together they form a unique fingerprint.

Cite this