論文アブストラクト： Our perception of our surrounding environment is limited by the constraints of human biology. The field of augmented perception asks how our sensory capabilities can be usefully extended through computational means. We argue that spatial awareness can be enhanced by exploiting recent advances in computer vision which make high-accuracy, real-time object detection feasible in everyday settings. We introduce HindSight, a wearable system that increases spatial awareness by detecting relevant objects in live 360-degree video and sonifying their position and class through bone conduction headphones. HindSight uses a deep neural network to locate and attribute semantic information to objects surrounding a user through a head-worn panoramic camera. It then uses bone conduction headphones, which preserve natural auditory acuity, to transmit audio notifications for detected objects of interest. We develop an application using HindSight to warn cyclists of approaching vehicles outside their field of view and evaluate it in an exploratory study with 15 users. Participants reported increases in perceived safety and awareness of approaching vehicles when using HindSight.
論文アブストラクト： Map users often need to interact repetitively with multiple important locations. For example, a traveler may frequently check her hotel or a train station on a map, use them to localize an unknown location, or investigate routes involving them. Ironically, these location-centric tasks cannot be performed using locations directly; users must instead pan and zoom the map or use a menu to access locations. We propose SpaceTokens, interactive widgets that act as clones of locations, and which users can create and place on map edges like virtual whiteboard magnets. SpaceTokens make location a first-class citizen of map interaction. They empower users to rapidly perform location-centric tasks directly using locations: users can select combinations of on-screen locations and SpaceTokens to control the map window, or connect them to create routes. Participants in a study overwhelmingly preferred a SpaceTokens prototype over Google Maps on identical smartphones for the majority of tasks.
論文アブストラクト： Ride-sharing can potentially address transportation challenges such as traffic congestion and air pollution by letting drivers share their cars unused capacity with a number of passengers. However, even though multiple ride-sharing services exist and HCI research has investigated various aspects of their use, we still have limited knowledge on how passengers use ride-sharing services to plan their trips. In this paper, we study how passengers use existing services to support the activity of planning a trip. We report from a qualitative study where we participated in 5 rides and conducted interviews with 19 passengers about their use and opinions towards ride-sharing services. We found that planning a ride involves comparing individual preferences across a number of services which enabled participants to support finding a trip and handle challenges such as privacy and trust. Further, we discuss these findings and their implications for future HCI research in ride-sharing.
論文アブストラクト： As humans, we have the natural capability of localizing the origin of sounds. Spatial audio rendering leverages this skill by applying special filters to recorded audio to create the impression that a sound emanates from a certain position in the physical space. A main application for spatial audio on mobile devices is to provide non-visual navigation cues. Current systems require users to either listen to artificial beacon sounds, or the entire audio source (e.g., a song) is repositioned in space, which impacts the listening experience. We present NavigaTone, a system that takes advantage of multi-track recordings and provides directional cues by moving a single track in the auditory space. While minimizing the impact of the navigation component on the listening experience, a user study showed that participants could localize sources as good as with stereo panning while the listening experience was rated to be closer to common music listening.