May 9, 2012 @ 09:30, Room: Ballroom G
Chair: Michael Haller, Media Interaction Lab, Austria
Rewarding the original: Explorations in joint user-sensor motion spaces - Paper
Contribution & Benefit: Describes a general technique to identify a set of communicative motions for a given input system by rewarding users for performing novel behaviours. Provides a systematic tool for designing gestures.
Abstract » This paper presents a systematic and general technique for
establishing a set of motions suitable for use with sensor
systems, by drawing performable and measurable motions
directly from users. It uses reinforcement which rewards
originality to induce users to explore the space of motions
they can perform. A decomposition of movements into motion
primitives is constructed, among which a meaningful
originality metric can be defined. Because the originality
measure is defined in terms of the sensed input, the resulting
space contains only movements which can both be performed
and sensed. We show how this can be used to evaluate
the relative performance of different joint user-sensor
systems, providing objective analyses of gesture lexicons
with regard to the technical limitations of sensors and humans.
In particular, we show how the space of motions
varies across the arm for a body-mounted inertial sensor.
ACMestablishing a set of motions suitable for use with sensor
systems, by drawing performable and measurable motions
directly from users. It uses reinforcement which rewards
originality to induce users to explore the space of motions
they can perform. A decomposition of movements into motion
primitives is constructed, among which a meaningful
originality metric can be defined. Because the originality
measure is defined in terms of the sensed input, the resulting
space contains only movements which can both be performed
and sensed. We show how this can be used to evaluate
the relative performance of different joint user-sensor
systems, providing objective analyses of gesture lexicons
with regard to the technical limitations of sensors and humans.
In particular, we show how the space of motions
varies across the arm for a body-mounted inertial sensor.
Vignette: Interactive Texture Design and Manipulation with Freeform Gestures for Pen-and-Ink Illustration - Paper
Contribution & Benefit: Presents a sketch-based application for interactive pen-and-ink illustration. The novel interaction and workflow enables to create a wide range of paintings easily and quickly, along with preserving personal artistic style.
Abstract » Vignette is an interactive system that facilitates texture creation in pen-and-ink illustrations. Unlike existing systems, Vignette preserves illustrators� workflow and style: users draw a fraction of a texture and use gestures to automatically fill regions with the texture. We currently support both 1D and 2D synthesis with stitching. Our system also has interactive refinement and editing capabilities to provide a higher level texture control, which helps artists achieve their desired vision. A user study with professional artists shows that Vignette makes the process of illustration more enjoyable and that first time users can create rich textures from scratch within minutes.
ACMInstructing People for Training Gestural Interactive Systems - Paper
Contribution & Benefit: Findings regarding the affect of kinematic instruction modality on training gestural interactive systems. Guideline for developers to collect training data for gesture recognition systems that achieve correctness and coverage.
Abstract » Entertainment and gaming systems such as the Wii and XBox Kinect have brought touchless, body-movement based interfaces to the masses. Systems like these enable the estimation of movements of various body parts from raw inertial motion or depth sensor data. However, the interface developer is still left with the challenging task of creating a system that recognizes these movements as embodying meaning. The machine learning approach for tackling this problem requires the collection of data sets that contain the relevant body movements and their associated semantic labels. These data sets directly impact the accuracy and performance of the gesture recognition system and should ideally contain all natural variations of the movements associated with a gesture. This paper addresses the problem of collecting such gesture datasets. In particular, we investigate the question of what is the most appropriate semiotic modality of instructions for conveying to human subjects the movements the system developer needs them to perform. The results of our qualitative and quantitative analysis indicate that the choice of modality has a significant impact on the performance of the learnt gesture recognition system; particularly in terms of correctness and coverage.
ACMMaking Gestural Input from Arm-Worn Inertial Sensors More Practical - Note
Contribution & Benefit: Gesture recognition requires complex computation and tedious user-training. We present an efficient recognition method that achieves accurate recognition with only a single calibration gesture from each user.
Abstract » Gestural input can greatly improve computing experiences away from the desktop, and has the potential to provide always-available access to computing. Specifically, accelerometers and gyroscopes worn on the arm (e.g., in a wristwatch) can sense arm gestures, enabling natural input in untethered scenarios. Two core components of any gesture recognition system are detecting when a gesture is occurring and classifying which gesture a person has performed. In previous work, accurate detection has required significant computation, and high-accuracy classification has come at the cost of training the system on a per-user basis. In this note, we present a gesture detection method whose computational complexity does not depend on the duration of the gesture, and describe a novel method for recognizing gestures with only a single example from a new user.
ACMClipoid: An Augmentable Short-Distance Wireless Toolkit for 'Accidentally Smart Home' Environments - Note
Contribution & Benefit: Our study is to understand how users utilize an augmentable wireless technology toolkit to upgrade their home environment. It provides a new way of enabling an 'accidentally smart home' environment.
Abstract » Unlike lab environments, the existing environment is not built for smart applications, but rather should be 'upgraded' to support new technologies. The result of this process is called the 'accidentally smart home'. We developed Clipoid, an augmentable wireless technology toolkit for supporting the development of an 'accidentally smart home' environment. We observed the real user context (static, moving) with Clipoid. We present a guideline for developing an augmentation toolkit, and identify human needs of close proximity physical interaction and multiple users-public platforms.
ACM