ShadowPuppets

Abstract

Pico projectors attached to mobile phones allow users to view phone content using a large display. However, to provide input to projector phones, users have to look at the device, diverting their attention from the projected image. Additionally, other collocated users have no way of interacting with the device.We present ShadowPuppets, a system that supports collocated interaction with mobile projector phones. Shadow-Puppets allows users to cast hand shadows as input to mobile projector phones. Most people understand how to cast hand shadows, which provide an easy input modality. Additionally, they implicitly support collocated usage, as nearby users can cast shadows as input and one user can see and understand another user's hand shadows.We describe the results of three user studies. The first study examines what hand shadows users expect will cause various effects. The second study looks at how users perceive hand shadows, examining what effects they think various hand shadows will cause. Finally, we present qualitative results from a study with our functional prototype and discuss design implications for systems using shadows as input. Our findings suggest that shadow input can provide a natural and intuitive way of interacting with projected interfaces and can support collocated collaboration.

PV-Tiles: Towards Closely-Coupled Photovoltaic and Digital Materials for Useful, Beautiful and Sustainable Interactive Surfaces

“…We provided each group with two position-adjustable LED lamps for making shadows on a wall or table. While prior work has suggested the use of shadow interactions for projected content [5] and more recently as a control mechanism for objects in the environment [4], this part of the workshop was designed to elicit potential shadow gestures that could be directly detected by the PV-Tiles built into surfaces or objects without other external instrumentation. Each group carried out three tasks, during which we captured photos and notes:…”

“…At its core is a microcontroller 5 that runs the algorithms used to detect gestures and sends images and text to the displays via a driver module. 5 Custom electronic switches allow energy from the OPV cells to be routed to an energy harvester module 5 , which charges a small lithium-ion battery when the interface is not being used to detect gestures.…”

Distributed interaction design

“…Cowan and Li [26] followed Wobbrock et al's [131] advice and followed their elicitation study with another study reversing the elicitation process. Their study's aim was to determine whether participants were able to identify the effect of an action on a computing system-i.e., the discoverability of an action.…”

“…To put Crowdlicit through its paces, I conducted a distributed elicitation study based on Morris's lab-based "Web on the Wall" elicitation study [26]. The Crowdlicit study had 78 participants recruited from Amazon's Mechanical Turk (mTurk).…”

“…This chapter contributes the following: (1) the Crowdlicit system; (2) the new end-user identification method, which evaluates the identifiability of elicitation-study results; and (3) the empirical results of two studies-(i) a distributed elicitation study of gesture and voice commands for a web browser, based on prior work [26], and (ii)…”