This paper examines key challenges in supporting passenger use of augmented and virtual reality headsets in transit. These headsets will allow passengers to break free from the restraints of physical displays placed in constrained environments such as cars, trains and planes. Moreover, they have the potential to allow passengers to make better use of their time by making travel more productive and enjoyable, supporting both privacy and immersion. However, there are significant barriers to headset usage by passengers in transit contexts. These barriers range from impediments that would entirely prevent safe usage and function (e.g. motion sickness) to those that might impair their adoption (e.g. social acceptability). We identify the key challenges that need to be overcome and discuss the necessary resolutions and research required to facilitate adoption and realize the potential advantages of using mixed reality headsets in transit.

This paper presents a powerful approach to evaluating public technologies by capturing and analysing pedestrian traffic using computer vision. This approach is highly flexible and scales better than traditional ethnographic techniques often used to evaluate technology in public spaces. This technique can be used to evaluate a wide variety of public installations and the data collected complements existing approaches. Our technique allows behavioural analysis of both interacting users and non-interacting passers-by. This gives us the tools to understand how technology changes public spaces, how passers-by approach or avoid public technologies, and how different interaction styles work in public spaces. In the paper, we apply this technique to two large public displays and a street performance. The results demonstrate how metrics such as walking speed and proximity can be used for analysis, and how this can be used to capture disruption to pedestrian traffic and passer-by approach patterns.

No abstract

There may be differences between this version and the published version. You are advised to consult the publisher's version if you wish to cite from it.

Acoustic levitation enables a radical new type of humancomputer interface composed of small levitating objects. For the first time, we investigate the selection of such objects, an important part of interaction with a levitating object display. We present Point-and-Shake, a mid-air pointing interaction for selecting levitating objects, with feedback given through object movement. We describe the implementation of this technique and present two user studies that evaluate it. The first study found that users could accurately (96%) and quickly (4.1s) select objects by pointing at them. The second study found that users were able to accurately (95%) and quickly (3s) select occluded objects. These results show that Pointand-Shake is an effective way of initiating interaction with levitating object displays.

There may be differences between this version and the published version. You are advised to consult the publisher's version if you wish to cite from it.

We demonstrate a technique for rendering textured haptic surfaces in mid-air, using an ultrasound haptic display. Our technique renders tessellated 3D 'haptic' shapes with di erent waveform properties, creating surfaces with distinct perceptions.

Room-scale Virtual Reality (VR) has become an affordable consumer reality, with applications ranging from entertainment to productivity. However, the limited physical space available for room-scale VR in the typical home or office environment poses a significant problem. To solve this, physical spaces can be extended by amplifying the mapping of physical to virtual movement (translational gain). Although amplified movement has been used since the earliest days of VR, little is known about how it influences reach-based interactions with virtual objects, now a standard feature of consumer VR. Consequently, this paper explores the picking and placing of virtual objects in VR for the first time, with translational gains of between 1x (a one-to-one mapping of a 3.5m*3.5m virtual space to the same sized physical space) and 3x (10.5m*10.5m virtual mapped to 3.5m*3.5m physical). Results show that reaching accuracy is maintained for up to 2x gain, however going beyond this diminishes accuracy and increases simulator sickness and perceived workload. We suggest gain levels of 1.5x to 1.75x can be utilized without compromising the usability of a VR task, significantly expanding the bounds of interactive room-scale VR.