Hiroaki Tanie

Embodied Symbol Emergence Based on Mimesis Theory

“Mimesis” theory focused in the cognitive science field and “mirror neurons” found in the biology field show that the behavior generation process is not independent of the behavior cognition process. The generation and cognition processes have a close relationship with each other. During the behavioral imitation period, a human being does not practice simple joint coordinate transformation, but will acknowledge the parents’ behavior. It understands the behavior after abstraction as symbols, and will generate its self-behavior. Focusing on these facts, we propose a new method which carries out the behavior cognition and behavior generation processes at the same time. We also propose a mathematical model based on hidden Markov models in order to integrate four abilities: (1) symbol emergence; (2) behavior recognition; (3) self-behavior generation; (4) acquiring the motion primitives. Finally, the feasibility of this method is shown through several experiments on a humanoid robot.

Keyframe compression and decompression for time series data based on the continuous hidden Markov model

Memory of motion patterns as data, comparison of a new motion pattern with data, and playback of one from the data are inevitably involved in the information processing of intelligent robot systems. Such computation forms the computational foundation of learning, acquisition, recognition, and generation process of intelligent robotic systems. In this paper, we propose to apply the continuous hidden Markov model to establish the computational foundation, using which one obtains the specified number of keyframes and their probability distributions. The keyframes are optimally selected to maximize the likelihood. The probability distributions are to be used to compute comparison and playback. The proposed method is applied to the motion data of a humanoid robot as well as the time series image data, and its validity is to be discussed.

High Marker Density Motion Capture by Retroreflective Mesh Suit

This paper presents a method for capturing detailed human motion by using a suit covered with retroreflective mesh. We can attach huge number of markers on the subject without replacing the hardware of current passive optical motion capture systems. Compared to normal motion capture using spherical markers, the connectivity information of the mesh can be used to improve the efficiency and accuracy of the reconstruction process. As a result, the system can achieve faster and more precise measurement of hundreds of markers on the subject than other approaches such as using natural image or 3D scanning. The total computation time required to reconstruct the 3D mesh information including 408 markers (intersections) is 65.5 ms, allowing realtime motion capture at 15 fps.

Key feature extraction for probabilistic categorization of human motion patterns

Mimesis is a hypothesis that human intelligence originated where motion recognition and motion generation interact through imitation. We previously proposed the mathematical model of mimesis using hidden Markov models (HMM) and constructed the proto symbol space from parameters of each HMM. The proto symbol space included only 10 motion patterns. No attention was paid on the relationship between behavior pattern and parts of body. It is common that a human observer pays an attention to the relationship between the parts of body and the behaviors recognizing performers behavior pattern. In this paper, we discuss key feature extraction from a rich database of behavior patterns based on probabilistic categorization among HMMs. The method is also applied to extract body parts that characterize behavior patterns

A Statistic Model of Embodied Symbol Emergence

Mimesis theory is one of the primitive skill of imitative learning which is regarded as an origin of human intelligence because imitation is fundamental function for communication and symbol manipulation When the mimesis is adopted as learning method for humanoids loads for designing full body behavior would be decrease because bottom up learning approaches from robot side and top down teaching approaches from user side involved each other. Therefore we propose a behavior acquisition and understanding system for humanoids based on the mime sis theory. This system is able to abstract observed others behaviors into symbols to recognize others behavior using the symbols and to generate motion patterns using the symbols. In this paper we extend the mimesis model to geometric symbol space which contains relative distance information among symbols We also discuss how to generate complex behavior by geometric symbol manipulation in the symbol space and how to recognize novel behavior using combination of symbols by known symbols.