The Experts below are selected from a list of 297 Experts worldwide ranked by ideXlab platform
Tomohiro Shibata - One of the best experts on this subject based on the ideXlab platform.
-
Interpolation and extrapolation of motion patterns in the Proto-symbol space
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2008Co-Authors: Tetsunari Inamura, Tomohiro ShibataAbstract:We propose a new method for interpolation and extrapolation of motion patterns in Proto-symbol spaces. The Proto-symbol space is a topological space which abstracts motion patterns by utilizing continuous hidden Markov models, and the mimesis model that recognizes/generates known/unknown motion patterns by using this topological space. An interpolation algorithm for the Proto-symbol space has already been proposed, but it had a mathematical problem. Furthermore, extrapolation of motion patterns was not defined, and could not be calculated. In the new method, the synthesis of Proto-symbols is done separately for state transition probabilities and output probabilities, and the synthesis of the state transition probabilities is done in the time domain. Experiments in a simulation environment demonstrate the feasibility of this method. © 2008 Springer-Verlag Berlin Heidelberg.
-
Geometric Proto-symbol manipulation towards language-based motion pattern synthesis and recognition
2008 IEEE RSJ International Conference on Intelligent Robots and Systems IROS, 2008Co-Authors: Tetsunari Inamura, Tomohiro ShibataAbstract:In this paper, we propose an improved mimesis method for interpolation and extrapolation of motion patterns in the Proto-symbol space towards an ultimate goal that motion pattern synthesis and recognition of humanoid robots are achieved by means of natural language. The Proto-symbol space is a topological space which abstracts motion patterns by utilizing continuous hidden Markov models. An interpolation algorithm for the Proto-symbol space was proposed in a previous work, but an extrapolation algorithm was not. Therefore, in this study, we propose and extrapolation method which can further clarify the physical meaning of the dimension of the Proto-symbol space that is one of the most essential issues for the realization of translation between motion patterns and language using the Proto-symbol space. The extrapolation method also enables the robot to recognize and synthesis various kinds of motion patterns using a fewer number of Protosymbols. The feasibility of the proposed method is demonstrated through simulation experiments. ©2008 IEEE.
Tetsunari Inamura - One of the best experts on this subject based on the ideXlab platform.
-
Interpolation and extrapolation of motion patterns in the Proto-symbol space
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2008Co-Authors: Tetsunari Inamura, Tomohiro ShibataAbstract:We propose a new method for interpolation and extrapolation of motion patterns in Proto-symbol spaces. The Proto-symbol space is a topological space which abstracts motion patterns by utilizing continuous hidden Markov models, and the mimesis model that recognizes/generates known/unknown motion patterns by using this topological space. An interpolation algorithm for the Proto-symbol space has already been proposed, but it had a mathematical problem. Furthermore, extrapolation of motion patterns was not defined, and could not be calculated. In the new method, the synthesis of Proto-symbols is done separately for state transition probabilities and output probabilities, and the synthesis of the state transition probabilities is done in the time domain. Experiments in a simulation environment demonstrate the feasibility of this method. © 2008 Springer-Verlag Berlin Heidelberg.
-
Geometric Proto-symbol manipulation towards language-based motion pattern synthesis and recognition
2008 IEEE RSJ International Conference on Intelligent Robots and Systems IROS, 2008Co-Authors: Tetsunari Inamura, Tomohiro ShibataAbstract:In this paper, we propose an improved mimesis method for interpolation and extrapolation of motion patterns in the Proto-symbol space towards an ultimate goal that motion pattern synthesis and recognition of humanoid robots are achieved by means of natural language. The Proto-symbol space is a topological space which abstracts motion patterns by utilizing continuous hidden Markov models. An interpolation algorithm for the Proto-symbol space was proposed in a previous work, but an extrapolation algorithm was not. Therefore, in this study, we propose and extrapolation method which can further clarify the physical meaning of the dimension of the Proto-symbol space that is one of the most essential issues for the realization of translation between motion patterns and language using the Proto-symbol space. The extrapolation method also enables the robot to recognize and synthesis various kinds of motion patterns using a fewer number of Protosymbols. The feasibility of the proposed method is demonstrated through simulation experiments. ©2008 IEEE.
Jianming Zheng - One of the best experts on this subject based on the ideXlab platform.
-
the earliest high fired glazed ceramics in china the composition of the Proto porcelain from zhejiang during the shang and zhou periods c 1700 221 bc
Journal of Archaeological Science, 2011Co-Authors: Thilo Rehren, Jianming ZhengAbstract:Abstract Bodies and glazes of 54 Proto-porcelain sherds and 18 non Proto-porcelain samples from Shang and Zhou periods production sites in Deqing, Zhejiang province were analysed by EPMA-WDS. The results indicate that the bodies of all samples were made from local raw material – porcelain stone, with the Proto-porcelain samples being made from clay of higher quality. Wood ashes, high in lime and low in potash, were intentionally applied to the Proto-porcelain samples, resulting in the formation of lime-rich glazes whose compositions were determined by a temperature-controlled mechanism. In contrast, kiln fragments and furniture show a potash-rich fuel vapour glaze, which formed unintentionally during use of the kiln. The firing temperature for most of the Proto-porcelain glazes is around the maturing temperature for typical more recent lime glazes, showing that the potters were already at such an early time able to attain sufficiently high temperature in their kilns.
Zhidong Li - One of the best experts on this subject based on the ideXlab platform.
-
Visual tracking by Proto-objects
Pattern Recognition, 2013Co-Authors: Zhidong Li, Weihong Wang, Yang Wang, Fang Chen, Yi WangAbstract:In this paper, we propose a biologically inspired framework of visual tracking based on Proto-objects. Given an image sequence, Proto-objects are first detected by combining saliency map and topic model. Then the target is tracked based on spatial and saliency information of the Proto-objects. In the proposed Bayesian approach, states of the target and Proto-objects are jointly estimated over time. Gibbs sampling has been used to optimize the estimation during the tracking process. The proposed method robustly handles occlusion, distraction, and illumination change in the experiments. Experimental results also demonstrate that the proposed method outperforms the state-of-the-art methods in challenging tracking tasks.
-
Saliency detection based on Proto-objects and topic model
2011 IEEE Workshop on Applications of Computer Vision (WACV), 2011Co-Authors: Zhidong Li, Yang Wang, Jie Xu, Glenn Geers, Jun YangAbstract:This paper proposes a novel computational framework for saliency detection, which integrates the saliency map computation and Proto-objects detection. The Proto-objects are detected based on the saliency map using latent topic model. The detected Proto-objects are then utilized to improve the saliency map computation. Extensive experiments are performed on two publicly available datasets. The experimental results show that the proposed framework outperforms the state-of-art methods.
Yueh-ning Lee - One of the best experts on this subject based on the ideXlab platform.
-
Formation of Proto-cluster: a virialized structure from gravo-turbulent collapse II. A two-dimensional analytical model for rotating and accreting system
Astronomy & Astrophysics, 2016Co-Authors: Yueh-ning LeeAbstract:Most stars are born in the gaseous Proto-cluster environment. The knowledge of this intermediate stage gives more accurate constraints on star formation characteristics. We demonstrate that a virialized globally supported structure, in which star formation happens, is formed out of a collapsing molecular cloud, and derive a mapping from the parent cloud parameters to the Proto-cluster to predict its properties, with a view to confront analytical calculations with observations and simulations. The virial theorem is decomposed into two dimensions to account for the rotation and the flattened geometry. Equilibrium is found by balancing rotation, turbulence and self-gravity, while turbulence is maintained by accretion driving and dissipates in one crossing time. The angular momentum and the accretion rate of the Proto-cluster are estimated from the parent cloud properties. The two-dimensional virial model predicts the size and velocity dispersion given the mass of the Proto-cluster and that of the parent cloud. The gaseous Proto-clusters lie on a sequence of equilibrium with the trend $R \sim M^{0.5}$, with limited variations depending on the evolutionary stage, the parent cloud, and the parameters not well known like turbulence driving efficiency by accretion and the turbulence anisotropy. The model reproduces successfully observations and simulation results. The properties of Proto-clusters follow universal relations and they can be derived from that of the parent cloud. Using simple estimates to infer the peak position of the core mass function (CMF) we find a weak dependence on the cluster mass suggesting that the physical conditions inside Proto-clusters may contribute to set a CMF, and by extension an IMF, that looks independent of the environment.
