Semantic action recognition by learning a pose lexicon

Lijuan Zhou, Wanqing Li, Philip Ogunbona, Zhengyou Zhang

Abstract: Highlights•A novel semantic representation, pose lexicon, is proposed for action recognition.•An extended hidden Markov alignment model is developed to learn a pose lexicon.•A semantic action recognition method that is capable of zero-shot recognition is developed upon the lexicon.•The efficacy of the proposed learning and recognition algorithms were evaluated on five datasets using cross-subject, cross-dataset and zero-shot protocols.