Marten Jeroen Pijl

An event-based approach to multi-modal activity modeling and recognition

The topic of human activity modeling and recognition still provides many challenges, despite receiving considerable attention. These challenges include the large number of sensors often required for accurate activity recognition, and the need for user-specific training samples. In this paper, an approach is presented for recognition of activities of daily living (ADL) using only a single camera and microphone as sensors. Scene analysis techniques are used to classify audio and video events, which are used to model a set of activities using hidden Markov models. Data was obtained through recordings of 8 participants. The events generated by scene analysis algorithms are compared to events obtained through manual annotation. In addition, several model parameter estimation techniques are compared. In a number of experiments, it is shown that if activities are fully observed these models yield a class accuracy of 97% on annotated data, and 94% on scene analysis data. Using a sliding window approach to classify activities in progress yields a class accuracy of 79% on annotated data, and 73% on scene analysis data. It is also shown that a multi-modal approach yields superior results compared to either individual modality on scene analysis data. Finally, it can be concluded the created models perform well even across participants.

Predicting daily physical activity in a lifestyle intervention program

The critical temperature of glass surface for condensation of mercury vapor as a function of density of atoms striking the surface has been found. The results indicate a large nucleus composed of many atoms is necessary for condensation to start; the heat of vaporization of an atom from the nucleus is 13.2 kcal. per mole. Some qualitative observations are reported, showing both rapid and slow adsorption of mercury atoms on glass and metal, and a considerable adsorption of mercury ions.

The Role of Empathy in Making Availability Judgments from Video and Silhouette Awareness Information

This paper reports research concerning video media spaces for the home and specifically the extent to which different approaches for video obfuscation can balance conflicting requirements for awareness between connected individuals and privacy. Different filtering techniques were compared regarding the ability of observers to make inferences regarding the observed, and with regards to the acceptability of being observed through such media. So far, related research has only considered individual differences between observers at a cursory level. We report on two experiments that evaluated whether people ability to empathize with others influences their ability to evaluate the availability of another person based on a video footage. We focused on supporting judgments of availability for communication comparing full video and on silhouette based obfuscation. The first experiment indicated a strong relation between empathy score and availability judgments. This effect was strongest for both males and females in the silhouette visualization condition. To further understand and confirm the effects found in this study, a second experiment involving more test participants and controlling for the correctness of availability judgments was conducted. Our findings suggest that empathetic skills specifically, and social cognition skills more generally, are critical factors for availability judgments.