Bayesian Classification of Task-oriented Actions Based on Stochastic Context-free Grammar. Yamamoto, M., Mitomi, H., Fujiwara, F., & Sato, T. In International Conference on Automatic Face and Gesture Recognition, pages 317--322, 2006.
abstract   bibtex   
This paper proposes a new approach for recognition of task-oriented actions based on stochastic context-free grammar (SCFG). Our attention puts on actions in the Japanese tea ceremony, where the action can be described by context-free grammar. Our aim is to recognize the action in the tea services. Existing SCFG approach consists of generating symbolic string, parsing it and recognition. The symbolic string often includes uncertainty. Therefore, the parsing process needs to recover the errors at the entry process. This paper proposes a segmentation method errorless as much as possible to segment an action into a string of finer actions. This method, based on an acceleration of the body motion, can produce the fine action corresponding to a terminal symbol with little error. After translating the sequence of fine actions into a set of symbolic strings, SCFG-based parsing of this set leaves small number of ones to be derived. Among the remaining strings, Bayesian classifier answers the action name with a maximum posterior probability. Giving one SCFG rule the multiple probabilities, one SCFG can recognize multiple actions
@InProceedings{Yamamoto2006,
  author    = {Yamamoto, M. and Mitomi, H. and Fujiwara, F. and Sato, T.},
  title     = {Bayesian Classification of Task-oriented Actions Based on Stochastic Context-free Grammar},
  booktitle = {International Conference on Automatic Face and Gesture Recognition},
  year      = {2006},
  pages     = {317--322},
  abstract  = {This paper proposes a new approach for recognition of task-oriented actions based on stochastic context-free grammar (SCFG). Our attention puts on actions in the Japanese tea ceremony, where the action can be described by context-free grammar. Our aim is to recognize the action in the tea services. Existing SCFG approach consists of generating symbolic string, parsing it and recognition. The symbolic string often includes uncertainty. Therefore, the parsing process needs to recover the errors at the entry process. This paper proposes a segmentation method errorless as much as possible to segment an action into a string of finer actions. This method, based on an acceleration of the body motion, can produce the fine action corresponding to a terminal symbol with little error. After translating the sequence of fine actions into a set of symbolic strings, SCFG-based parsing of this set leaves small number of ones to be derived. Among the remaining strings, Bayesian classifier answers the action name with a maximum posterior probability. Giving one SCFG rule the multiple probabilities, one SCFG can recognize multiple actions},
  groups    = {Lit Review 2013-09, EMBC2014},
  review    = {Want to classify the Japanese tea ceremony. Want to use stochastic context free grammar (SCFG) to do so. It is relevant to temae (a set of rules for the tea ceremony, and consists of sub-actions). They segment by looking at local minimums on the sum of the magnitude of the angular acceleration.


Yamamoto \etal \cite{Yamamoto2006} want to segment and classify the different components of the Japanese tea ceremony. The state progress is presented by the stochastic context free grammar and auto-regression models, but the segmentation is performed by examining local minimums on the sum of the magnitude of the angular acceleration. No segmentation accuracy was given.

Yamamoto \etal \cite{Yamamoto2006} segment and classify the different components of the Japanese tea ceremony via image-based segmentation. The segmentation is performed by examining local minimums on the sum of the magnitude of the angular acceleration in the actor poses. Labels are assigned via a SCFG.},
  timestamp = {2013.10.08},
}
Downloads: 0