Kohjiro Hashimoto

Modeling method of human actions based on the situation with variable level of the spatial-temporal abstraction

In order to realize a system that adapts to a person by considering human behaviors such as an automatic monitoring system and a driving support system, the system must have a certain model of human behaviors. Therefore, we have proposed a modeling method of human behavior based on the causality between a situation around a person and a human behavior. In order to model human behaviors appropriately depending on the problem where the proposed modeling method is applied, the spatial and temporal abstraction levels of human behaviors should be adjustable. In this paper, we propose a modeling method of human behaviors that can adjust the spatial and temporal abstraction levels of modeled human behaviors. In this method, the temporal and spatial abstraction levels of human behavior can be adjust independently, and the spatial and temporal adjustment parameters are introduced in the time series clustering process. The usefulness of the proposed method is examined through the experiment.

Statistical modeling method of human actions expressed by multi-dimentional time series data with Hidden Markov Model

In this paper, a modeling method of human actions is proposed in order to realize such systems as to assist human operations have been desired, which must have a certain human action model to recognize or support various kinds of human actions. In the proposed method, a human action model is extracted statistically from enormous data obtained by long-term observation of human actions with sensors, which means only frequent human actions are modeled in this method. In addition, the human action model obtained by the proposed method has high readability, which makes human action analysis much easier. In order to generate a human action model with the previous two features, a human action and a situation around a person are modeled as time series data expressed by Hidden Markov Model(HMM). This is because HMM can efficiently model a time series data with temporal and spatial redundancy. In addition, the relationship between a situation and a human action modeled by HMMs is expressed by If-Then-Rule style explicitly.

Estimation of next human action and its timing based on the human action model with timing probabilistic distribution

In order to give a suitable support to a person timely, it is necessary for the system to estimate the next human action and its execution timing. Therefore, we have proposed a modeling method of human actions based on the causality between the situation around a person and human action change. Our previous modeling method makes it possible to estimate the next human action and its execution timing based on Hidden Markov Model which expresses the situation around a person. However, Hidden Markov Model was not enough to express the temporal information such as execution timing of next human action. According to this reason, we propose a new model structure of human action that a probabilistic distribution of timing information is incorporated into Hidden Markov Model. And we propose a new modeling method of human action based on the above new probabilistic model.

Estimation of execution timing of human action considering causality between action and situation

In order to realize a system to support human actions, the system should be able to estimate the next human action. Therefore, we have proposed a modeling method of human actions based on the causality between the situation around a person and the human action change. Our previous modeling method makes it possible to estimate the next human action. However, in order to give a suitable support to a person, it is necessary for the system to estimate not only the next human action, but also its execution timing. In this paper, we propose a modeling method of human actions for estimate the execution timing of the next human action by expanding our previous method. In addition, an early estimation method of the next human action and its execution timing is also proposed.

Estimation of next human behavior and its timing for human behavior support

We have proposed a modeling and recognition method of human behaviors in this research. In the proposed modeling method, we have assumed that a person changes his behavior according to the change of the situation around him, and this concept is expressed by If-Then-Rules, which are called behavior rules. In behavior rules, the change of the situation around a person is described by Hidden Markov Model(HMM) which models multi-dimensional time series sensing data. In this research, a support system for human driving behaviors has been assumed as an example of application of the proposed model. In order to realize the system which shows a driver his next behavior, we propose an estimation method of the next human behavior and the timing of its execution. The usefulness of the proposed method is examined through experimental results of behavior recognition with the constructed system.

Extraction of human action elements with transition network of partial time series data modeled by Hidden Markov Model

The authors have researched on a method of human action modeling to realize systems such as to support human human operations or watch persons to prevent various kinds of accidents. In order to recognize or support various kinds of human actions, a certain human action model is necessary in these systems. Therefore, we have proposed a modeling method of human actions, which is extracted statistically from enormous data acquired from various kinds of sensors by long-term observation of human actions and situations around persons. However, human action elements composing a model should have been extracted heuristically by a designer. In this paper, an extraction method of human action elements is proposed in order to extract frequent human action elements automatically from acquired data. In the proposed method, a series of acquired time series data is divided into short partial ones, which are modeled by Hidden Markov Models(HMM). Then, the transition network of generated HMMs is constructed based on the likelihood between the original data and each HMM. In the obtained network, a transition sequence with the only one edge is regarded as a frequent human action element. Extraction results with artificial and actual human action data are shown in this paper in order to verify the usefulness of the proposed method.

Modeling method of execution timing of operation to analyze the reaction time from judgment to execution of operation

In this study, we aim to evaluate a driver skill based on the driver behavior model generated by the operation and environment data under driving vehicle. When skill of driver is reduced, the operator behavior is delayed, it takes along time of operation, and execution timing of operation is no reproducibility. These mean that reducing skill of driver affects time information of the driver operation. Therefore, we study on the model structure incorporated time information expresses the execution timing of drivers operation to evaluate drivers skill. In particular, we propose a new model structure incorporated timing probabilistic distribution expresses the time until the execution of next operation. Moreover, we propose a modeling method of operator behavior according to proposed model structure. It is considered that the reaction time from the decision to execution of operation and ambiguity of the decision can be read from the timing probabilistic distribution. In this paper, the above usefulness is examined through some experimental results. Moreover, the applicability of the driving skill evaluation based on the proposed operation model is examined.

Modeling method of human action with HS considering its temporal and spatial differences

In this paper, a modeling method of human actions is proposed. This is because systems to assist human operations have been desired, which must have a certain human action model to recognize or support various kinds of human actions. In the proposed method, a human action model is statistically generated by extracted from enormous data obtained by long-term monitoring with sensors. Therefore, human actions can be modeled without previous knowledge. In addition, a human action model is generated considering the differences on not only spatial patterns but also temporal ones of human actions. This is because it is very significant to support human operations timely. In order to generate a human action model with the previous two features, a human action is modeled as a human action pattern expressed by Hidden Semi Markov Model(HSMM) and a If-Then-Rule in the proposed method. HSMM can model a time series data explicitly. It can also model the timing information of a change of a human action. In addition, the relationship between the situation and a human action is expressed by If-Then-Rule explicitly. Therefore, a human action model has high readability.

Modeling and recognition method of human behaviors with multi-dimensional time series data

We propose a modeling and recognition method of human behaviors in this paper in order to realize such intelligent systems that can adapt humans, i.e. the systems that support humans by considering human behaviors, In the proposed method, we assume that a human changes his behavior according to the change of the situation around him, and this concept is expressed by If-Then-Rules, which is called behavior rules. In behavior rules, the change of the situation around a person is described by multi-dimensional time series sensing data which is modeled with Hidden Markov Model(HMM). To recognize the change of human behaviors, the optimal If-Then-Rule is chosen based on the current human behavior and similarity to the time series data of the situation obtained by sensors. As an example of human behaviors, human driving behaviors are considered, and a recognition system of human driving behaviors is constructed. The usefulness of the proposed method is examined through some experimental results with the constructed system.