@InProceedings{Janus2005,
  Title                    = {Unsupervised Probabilistic Segmentation of Motion Data for Mimesis Modeling},
  Author                   = {Janus, B. and Nakamura, Y.},
  Booktitle                = {Proceedings of the IEEE International Conference on Advanced Robotics},
  Year                     = {2005},
  Pages                    = {411--417},

  Abstract                 = {Humanoid developments express the need for intelligent learning systems that can automatically realize behavior acquisition and symbol emergence. In the framework of mimesis model, we present an unsupervised dynamic HMM-based algorithm in order to analyze vectorial motion data. The efficiency of this algorithm is demonstrated by segmenting continuous sequence of real movements. We also propose to use it as the first level of an information treatment system by associating it with a recognition process. Unlike other existing segmentation-recognition system, our segmentation process does not need any learning of the parameters that increases the flexibility of the whole segmentation-recognition system and the range of its possible applications},
  Doi                      = {10.1109/ICAR.2005.1507443},
  Keywords                 = {behavior acquisition;information treatment system;intelligent learning systems;mimesis modeling;motion data segmentation;segmentation-recognition system;symbol emergence;unsupervised dynamic HMM-based algorithm;unsupervised probabilistic segmentation;hidden Markov models;humanoid robots;image motion analysis;image recognition;image segmentation;learning systems;probability;},
  Review                   = {Janus and Nakamura \cite{Janus2005} apply Kohlmorgen and Lemm's algorithm \cite{Kohlmorgen2002} to human movement data. HMM state changes occurred when the signal's distribution function is sufficiently different from the previous state, suggesting possible segmentation points at these state changes. This algorithm was verified on a dataset of a human subject performing various different full body exercises, captured with motion capture and translated to joint angles via inverse kinematics. However, segmentation accuracy was not provided.},
  Timestamp                = {2011.04.18}
}

{"_id":"r7J8kpySWFmeZC6af","bibbaseid":"janus-nakamura-unsupervisedprobabilisticsegmentationofmotiondataformimesismodeling-2005","downloads":0,"creationDate":"2017-09-14T16:34:36.571Z","title":"Unsupervised Probabilistic Segmentation of Motion Data for Mimesis Modeling","author_short":["Janus, B.","Nakamura, Y."],"year":2005,"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/jfslin/jfslin.github.io/master/jf2lin.bib","bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Unsupervised Probabilistic Segmentation of Motion Data for Mimesis Modeling","author":[{"propositions":[],"lastnames":["Janus"],"firstnames":["B."],"suffixes":[]},{"propositions":[],"lastnames":["Nakamura"],"firstnames":["Y."],"suffixes":[]}],"booktitle":"Proceedings of the IEEE International Conference on Advanced Robotics","year":"2005","pages":"411--417","abstract":"Humanoid developments express the need for intelligent learning systems that can automatically realize behavior acquisition and symbol emergence. In the framework of mimesis model, we present an unsupervised dynamic HMM-based algorithm in order to analyze vectorial motion data. The efficiency of this algorithm is demonstrated by segmenting continuous sequence of real movements. We also propose to use it as the first level of an information treatment system by associating it with a recognition process. Unlike other existing segmentation-recognition system, our segmentation process does not need any learning of the parameters that increases the flexibility of the whole segmentation-recognition system and the range of its possible applications","doi":"10.1109/ICAR.2005.1507443","keywords":"behavior acquisition;information treatment system;intelligent learning systems;mimesis modeling;motion data segmentation;segmentation-recognition system;symbol emergence;unsupervised dynamic HMM-based algorithm;unsupervised probabilistic segmentation;hidden Markov models;humanoid robots;image motion analysis;image recognition;image segmentation;learning systems;probability;","review":"Janus and Nakamura i̧teJanus2005 apply Kohlmorgen and Lemm's algorithm i̧teKohlmorgen2002 to human movement data. HMM state changes occurred when the signal's distribution function is sufficiently different from the previous state, suggesting possible segmentation points at these state changes. This algorithm was verified on a dataset of a human subject performing various different full body exercises, captured with motion capture and translated to joint angles via inverse kinematics. However, segmentation accuracy was not provided.","timestamp":"2011.04.18","bibtex":"@InProceedings{Janus2005,\n Title = {Unsupervised Probabilistic Segmentation of Motion Data for Mimesis Modeling},\n Author = {Janus, B. and Nakamura, Y.},\n Booktitle = {Proceedings of the IEEE International Conference on Advanced Robotics},\n Year = {2005},\n Pages = {411--417},\n\n Abstract = {Humanoid developments express the need for intelligent learning systems that can automatically realize behavior acquisition and symbol emergence. In the framework of mimesis model, we present an unsupervised dynamic HMM-based algorithm in order to analyze vectorial motion data. The efficiency of this algorithm is demonstrated by segmenting continuous sequence of real movements. We also propose to use it as the first level of an information treatment system by associating it with a recognition process. Unlike other existing segmentation-recognition system, our segmentation process does not need any learning of the parameters that increases the flexibility of the whole segmentation-recognition system and the range of its possible applications},\n Doi = {10.1109/ICAR.2005.1507443},\n Keywords = {behavior acquisition;information treatment system;intelligent learning systems;mimesis modeling;motion data segmentation;segmentation-recognition system;symbol emergence;unsupervised dynamic HMM-based algorithm;unsupervised probabilistic segmentation;hidden Markov models;humanoid robots;image motion analysis;image recognition;image segmentation;learning systems;probability;},\n Review = {Janus and Nakamura \\cite{Janus2005} apply Kohlmorgen and Lemm's algorithm \\cite{Kohlmorgen2002} to human movement data. HMM state changes occurred when the signal's distribution function is sufficiently different from the previous state, suggesting possible segmentation points at these state changes. This algorithm was verified on a dataset of a human subject performing various different full body exercises, captured with motion capture and translated to joint angles via inverse kinematics. However, segmentation accuracy was not provided.},\n Timestamp = {2011.04.18}\n}\n\n","author_short":["Janus, B.","Nakamura, Y."],"key":"Janus2005","id":"Janus2005","bibbaseid":"janus-nakamura-unsupervisedprobabilisticsegmentationofmotiondataformimesismodeling-2005","role":"author","urls":{},"keyword":["behavior acquisition;information treatment system;intelligent learning systems;mimesis modeling;motion data segmentation;segmentation-recognition system;symbol emergence;unsupervised dynamic HMM-based algorithm;unsupervised probabilistic segmentation;hidden Markov models;humanoid robots;image motion analysis;image recognition;image segmentation;learning systems;probability;"],"downloads":0},"search_terms":["unsupervised","probabilistic","segmentation","motion","data","mimesis","modeling","janus","nakamura"],"keywords":["behavior acquisition;information treatment system;intelligent learning systems;mimesis modeling;motion data segmentation;segmentation-recognition system;symbol emergence;unsupervised dynamic hmm-based algorithm;unsupervised probabilistic segmentation;hidden markov models;humanoid robots;image motion analysis;image recognition;image segmentation;learning systems;probability;"],"authorIDs":[],"dataSources":["iCsmKnycRmHPxmhBd"]}