Proceedings. Fourth IEEE International Conference on Multimodal Interfaces
We present the use of layered probabilistic representations using Hidden Markov Models for perfor... more We present the use of layered probabilistic representations using Hidden Markov Models for performing sensing, learning, and inference at multiple levels of temporal granularity. We describe the use of the representation in a system that diagnoses states of a user's activity based on real-time streams of evidence from video, acoustic, and computer interactions. We review the representation, present an implementation, and report on experiments with the layered representation in an office-awareness application.
Proceedings of the 5th international conference on Multimodal interfaces - ICMI '03, 2003
Perceptual user interfaces promise modes of fluid computerhuman interaction that complement the m... more Perceptual user interfaces promise modes of fluid computerhuman interaction that complement the mouse and keyboard, and have been especially motivated in non-desktop scenarios, such as kiosks or smart rooms. Such interfaces, however, have been slow to see use for a variety of reasons, including the computational burden they impose, a lack of robustness outside the laboratory, unreasonable calibration demands, and a shortage of sufficiently compelling applications. We address these difficulties by using a fast stereo vision algorithm for recognizing hand positions and gestures. Our system uses two inexpensive video cameras to extract depth information. This depth information enhances automatic object detection and tracking robustness, and may also be used in applications. We demonstrate the algorithm in combination with speech recognition to perform several basic window management tasks, report on a user study probing the ease of using the system, and discuss the implications of such a system for future user interfaces.
Popular content in video sharing websites (e.g., YouTube) contains many duplicates. Most scholars... more Popular content in video sharing websites (e.g., YouTube) contains many duplicates. Most scholars define near-duplicate video clips (NDVC) as identical videos with variations on non-semantic features (e.g., image/audio quality), while a few others also include semantic features (different videos of similar content). However, it is unclear what exact features contribute to human perception of similar videos. In this paper, we present the results of a user study conducted with 217 users of video sharing websites. Findings confirm the relevance of both classes of features, but the exact role played by semantics on each instance of NDVC is still an open question. In most cases, participants had a preference for one video when compared to its NDVC and they were more tolerant to changes in the audio than in the video channel.
Exploring persuasive techniques for medication compliance
Mobile applications that incorporate persuasive techniques have recently been shown to have a pos... more Mobile applications that incorporate persuasive techniques have recently been shown to have a positive impact in helping their users achieve pre-defined wellness goals (e.g., keeping active, eating healthier, etc.). In this paper, we present MoviPill, a mobile phone based application that combines a set of persuasive techniques (i.e., social competition through a game, social support, virtual rewards and entertaining selfmonitoring) to help patients improve their levels of medication compliance. In a 6-week field study, 18 elders used a simplified version of MoviPill that included only the game component. Still, both their compliance levels and the precision of the drug intake time improved by 60% and 43% respectively, when compared to the baseline (i.e. their usual medication reminding tools). We plan on evaluating the additional persuasive techniques with further studies.
We present HealthGear, a real-time wearable system for monitoring, visualizing and analyzing phys... more We present HealthGear, a real-time wearable system for monitoring, visualizing and analyzing physiological signals. HealthGear consists of a set of non-invasive physiological sensors wirelessly connected via Bluetooth to a cell phone which stores, transmits and analyzes the physiological data, and presents it to the user in an intelligible way. In this paper, we focus on an implementation of HealthGear using a blood oximeter to monitor the user's blood oxygen level and pulse while sleeping. We also describe two different algorithms for automatically detecting sleep apnea events, and illustrate the performance of the overall system in a sleep study with 20 volunteers. Index Terms-Wearable physiological sensors and monitoring, sleep apnea, mobile devices, pattern recognition of physiological signals.
We present the use of layered probabilistic representations for modeling human activities, and de... more We present the use of layered probabilistic representations for modeling human activities, and describe how we use the representation to do sensing, learning, and inference at multiple levels of temporal granularity and abstraction and from heterogeneous data sources. The approach centers on the use of a cascade of Hidden Markov Models named Layered Hidden Markov Models (LHMMs) to diagnose states of a userÕs activity based on real-time streams of evidence from video, audio, and computer (keyboard and mouse) interactions. We couple these LHMMs with an expected utility analysis that considers the cost of misclassification. We describe the representation, present an implementation, and report on experiments with our layered architecture in a real-time office-awareness setting.
Intensive computations required for sensing and processing perceptual information can impose sign... more Intensive computations required for sensing and processing perceptual information can impose significant burdens on personal computer systems. We explore several policies for selective perception in SEER, a multimodal system for recognizing office activity that relies on a layered Hidden Markov Model representation. We review our efforts to employ expected-value-of-information (EVI) computations to limit sensing and analysis in a context-sensitive manner. We discuss an implementation of a one-step myopic EVI analysis and compare the results of using the myopic EVI with a heuristic sensing policy that makes observations at different frequencies. Both policies are then compared to a random perception policy, where sensors are selected at random. Finally, we discuss the sensitivity of ideal perceptual actions to preferences encoded in utility models about information value and the cost of sensing.
Uploads
Papers by Nuria Oliver