Pointing Gesture
Analysis of Spartacus
To accurately select the target device, Spartacus makes an assumption that when a user points her phone towards the target device, the target device will always observe the largest Doppler frequency shift. However, in practice, user gestures can be significantly diverse in terms of directional precision, velocity, and trajectory. Such diversities make accurately estimating the peak frequency shift significantly challenging.
In order to investigate what factors may affect the performance of Spartacus, we conducted experiments to fully understand the characteristics of pointing gestures of average users. Some fundamental questions that we would want to investigate include:
1. How diversely do users
point their phones, and how fast can a user point?
2. If the user points fast
enough, how often does the tar- get device observe the
highest frequency shift, thus the highest velocity, of the
gesture?
3. If we want to estimate
the frequency shifts, how much frequency- and time-domain
resolution do we need to successfully capture the peak
frequency shift inside of a gesture?
To answer these questions, we conducted experiments with 12 participants (6 females) and investigated their pointing gestures. To capture the gesture trajectories, we attached a video recorder on the ceiling right above the participants, and videotaped the entire experiment. The video we took recorded complete 2D trajectories of the gestures. Before doing the experiment, we briefed the participants on the idea of Spartacus, and let them to freely choose any natural gestures they wanted. During the experiment, each participant performed 10 gestures towards a target device 2m away from them, using a Galaxy Nexus phone. A red marker was attached to the participants’ hands for motion-tracking. After the experiment, we detected hand trajectories of the participants using image processing techniques.
Quantitative analysis of the users' gesture trajectories can
be found in our MobiSys paper.
This webpage contains the videos that we have recorded and
used in our analysis. To facilitate processing these videos,
we also attached the Matlab code of tracking the color marker
in the videos.
You could download the videos individually, or as one compressed
archive file (109MB).
Male 1 - Video (m4v format)
Male 2 - Video (m4v format)
Male 3 - Video (m4v format)
Male 4 - Video (m4v format)
Male 5 - Video (m4v format)
Male 6 - Video (m4v format)
Female 1 - Video (m4v format)
Female 2 - Video (m4v format)
Female 3 - Video (m4v format)
Female 4 - Video (m4v format)
Female 5 - Video (m4v format)