Jointly Learning Visual Poses and Pose Lexicon for Semantic Action Recognition

• Published in: IEEE transactions on circuits and systems for video technology Vol. 30; no. 2; pp. 457 - 467
• Main Authors: Zhou, Lijuan, Li, Wanqing, Ogunbona, Philip, Zhang, Zhengyou
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Alignment; Computational modeling; Conditional probability; Datasets; Expectation-maximization algorithms; Hidden Markov models; Integrated circuit modeling; Learning; Mapping; Markov chains; pose lexicon; Probabilistic logic; Recognition; Semantic action recognition; Semantics; Statistical analysis; Visual observation; visual pose; Visualization
• ISSN: 1051-8215; 1558-2205
• DOI: 10.1109/TCSVT.2019.2890829

A novel method for semantic action recognition through learning a pose lexicon is presented in this paper. A pose lexicon comprises a set of semantic poses, a set of visual poses, and a probabilistic mapping between the visual and semantic poses. This paper assumes that both the visual poses and mapping are hidden and proposes a method to simultaneously learn a visual pose model that estimates the likelihood of an observed video frame being generated from hidden visual poses, and a pose lexicon model establishes the probabilistic mapping between the hidden visual poses and the semantic poses parsed from textual instructions. Specifically, the proposed method consists of two-level hidden Markov models. One level represents the alignment between the visual poses and semantic poses. The other level represents a visual pose sequence, and each visual pose is modeled as a Gaussian mixture. An expectation-maximization algorithm is developed to train a pose lexicon. With the learned lexicon, action classification is formulated as a problem of finding the maximum posterior probability of a given sequence of video frames that follows a given sequence of semantic poses, constrained by the most likely visual pose and the alignment sequences. The proposed method was evaluated on MSRC-12, WorkoutSU-10, WorkoutUOW-18, Combined-15, Combined-17, and Combined-50 action datasets using cross-subject, cross-dataset, zero-shot, and seen/unseen protocols.