SonarWatch: Appropriating the Forearm as a Slider Bar

彬彬

Copyright is held by the author / owner(s).
SIGGRAPH Asia 2011, Hong Kong, China, December 12 – 15, 2011.
ISBN 978-1-4503-0807-6/11/0012
SonarWatch: Appropriating the Forearm as a Slider Bar
Rong-Hao Liangy Shu-Yang Lin Chao-Huai Su Kai-Yin Cheng Bing-Yu Chen De-Nian Yangy
National Taiwan University y Academia Sinica
1 introduction
Human bodies become an emerging type of human-computer interfaces
recently. Not only because our skin is a surface that is always
available and highly accessible, but also the sense of how our
body is configured in space allows us to accurately interact with our
bodies in an eye-free manner. Hence, this input method is suitable
to be applied on extending the interaction space of mobile devices
[Harrison et al. 2010] or providing more degrees-of-freedom for enhancing
gaming experiences such as kinect1. Nevertheless, since
the additional gesture detector may be obtrusive or not so portable
for users, this approach can hardly be applied in everyday life.
This work presents SonarWatch, a less-obtrusive, wearable gesture
detector. Unlike the device introduced by Nakatsuma et al. [2011],
we equipped an ultrasonic rangefinder and a capacitive touch sensor
on a wristwatch (Figure 1), thus a user’s forearm can be appropriated
as a readily available input surface for interacting with computers.
Several click and slide gestures are allowed to be performed
by the user. To demonstrate the possible applications and the effectiveness
of this implicit design, three applications are provided:
photo browsing, music playlist controlling and augmenting Kinect
for gaming.


Appropriating the Forearm as a Slider Bar
Huai Su1 Kai-Yin Cheng1 Bing-Yu Chen1 De-Nian Yang2
Networking and Multimedia, National Taiwan University
Information Science, Academia Sinica
computer interfaces
surface that is
our body
with our
suitable to be
devices [Harrison
enhancing
detector
life.
SonarWatch, a
capacitive
available
slide on an
operations
applications, photo
then developed
SonarWatch in our
facilitate bodysensing
its server.
gesture recognition
Devantech
readings
mounted for
connecting
Arduino Mini
Raw data
string is received,
procedure.
transferred to the
Capacitive Touch
Sensor
Ultrasonic
Rangefinder
Micro-processor
on Arduino Mini
Figure 1: SonarWatch, a wristwatch equipped with an ultrasonic
rangefinder and a capacitive touch sensor. Sensor outputs are retrieved
by the Arduino Mini platform
2.2 Gesture recognition
Our gesture recognition procedure is illustrated as Figure 2. When
user touch or pinch the capacitive touch sensor, the sensor would
output in high, which can be regarded as the “ready” signal.
Once the capacitive touch sensor is released and the ultrasonic
rangefinder reveals that no finger is within the 20cm range of detection,
a Click gesture (Figure 3(a)(b)) can be immediately determined.
Otherwise, the ultrasonic rangefinder would start sampling
in 15Hz in 1 second, and user is allowed to perform their
gesture in this period. In the end of sampling, a Slide gesture
can be determined. Advanced Click gestures such as Single-click
and Double-click; Slide gestures such as Slide-Up, Slide-Down,
and Slide-to-Select can also be determined by simple classification
(Figure 3(c)(d)(e)) on sensor values.
Figure 1: SonarWatch, a wristwatch equipped with an ultrasonic
rangefinder and a capacitive touch sensor. Sensor outputs are retrieved
by the Arduino Mini platform
2 SonarWatch
Our device consists of a low-power consumption Devantech SRF10
ultrasonic rangefinder, which provides distance readings from 6cm
to 6m, and a capacitive touch sensor mounted for detecting user
events. The user can touch the sensor to initiate the detection
(Figure 2(a)(b)). Once releasing the finger from the sensor, the
rangefinder starts detecting the gesture between the following 0:3s
to 1s at 30Hz sampling rate. Noisy outputs would be classified
as outliers. All sensor data is retrieved by the microprocessor and
transferred to the server running on a PC for analysis.
Two types of gestures are developed for users to interact with
SonarWatch: Click and Slide. A Click gesture (Figure 2(a)) is recognized
if there is no finger detected within 20cm in the first three
samplings. Once detecting a click, the sampling session would be
immediately terminated for listening a Double-Click. A Slide gesture
is recognized if fingers are detected within the range such like
1http://www.xbox.com/kinect/
(Figure 2(b)). In this case, the sampling session would be continued,
and the maximum distance M and the final position P would
be obtained. If M  20cm, Slide-to-Select (Figure 2(c)) is determined
with regarding selection at P; if M > 20cm, it is determined
as either Slide-Up (P > 20cm, Figure 2(d)) or Slide-Down
(P  20cm, Figure 2(e)).
Figure 3: Gestures supported by SonarWatch. (a) and (b) are Click
gesture. (c) and (d) are Slide-to-Select gesture. (c)and (e) are Slide-
Up gesture.
3 Application
We implement three different applications for demonstrating the effectiveness
of our approaches: photo browsing, playlist controlling,
and augmenting Kinect for gaming. The first two applications are
implemented by ActionScript 3.0 and Flash, and the third application
is implemented by OpenCV and OpenNI2.
3.1 Photo browsing
Rich gestures can help user browse photo more efficiently. In
our photo browsing application, user can Slide-Up/Down for turning
pages up/down, Slide-to-Select thumbnail for browsing photos,
Single-click to enter selected album, and Double-click to return.
3.2 Playlist controlling
User may want to control background application such as their
music player in an eyes-free manner. In playlist controlling application,
user can Single-click to play/pause the music, Slide-to-
Select to adjust volume, and Slide-Up/Down for switching to the
next/previous song.
3.3 Augmenting Kinect for gaming
Utilizing body as controller may somehow lack-of-haptic sensation.
Hence, our proprioception can be utilized as complement. We use
the Angry Birds3 game for illustration. By using Kinect to capture
user’s arm position, user can adjust the angle of bow. For launching,
user can Slide-to-Select the desired intensity, then remove their
finger to launch the Angry Birds.
2http://www.openni.org
3http://www.angrybirds.com/
4 Conclusion
We present SonarWatch, a less-obtrusive, wearable By equipping an ultrasonic rangefinder and sensor on the wristwatch, user’s forearm can be readily available input surface for interacting with click and slide gestures are allowed to be performed To demonstrate the possible applications and the this approach, three applications: photo browsing, controlling and augmenting Kinect for gaming, the SonarWatch. We believe that this work can facilitate technologies in our daily life.
References
HARRISON, C., TAN, D., AND MORRIS, D. Skinput: the body as an input surface. In Proc. ACM 462.
Figure 3: Gestures supported by SonarWatch. (a) and (b) are Click
gesture. (c) and (d) are Slide-to-Select gesture. (c)and (e) are Slide-
Up gesture.
3 Application
We implement three different applications for demonstrating the effectiveness
of our approaches: photo browsing, playlist controlling,
and augmenting Kinect for gaming. The first two applications are
implemented by ActionScript 3.0 and Flash, and the third application
is implemented by OpenCV and OpenNI2.
3.1 Photo browsing
Rich gestures can help user browse photo more efficiently. In
our photo browsing application, user can Slide-Up/Down for turning
pages up/down, Slide-to-Select thumbnail for browsing photos,
Single-click to enter selected album, and Double-click to return.
3.2 Playlist controlling
User may want to control background application such as their
music player in an eyes-free manner. In playlist controlling application,
user can Single-click to play/pause the music, Slide-to-
Select to adjust volume, and Slide-Up/Down for switching to the
next/previous song.
3.3 Augmenting Kinect for gaming
Utilizing body as controller may somehow lack-of-haptic sensation.
Hence, our proprioception can be utilized as complement. We use
the Angry Birds3 game for illustration. By using Kinect to capture
user’s arm position, user can adjust the angle of bow. For launching,
user can Slide-to-Select the desired intensity, then remove their
finger to launch the Angry Birds.
2http://www.openni.org
3http://www.angrybirds.com/
4 Conclusion
We present SonarWatch, a less-obtrusive, wearable gesture detector.
By equipping an ultrasonic rangefinder and a capacitive touch
sensor on the wristwatch, user’s forearm can be appropriated as a
readily available input surface for interacting with computers. Several
click and slide gestures are allowed to be performed by users.
To demonstrate the possible applications and the effectiveness of
this approach, three applications: photo browsing, music playlist
controlling and augmenting Kinect for gaming, are developed for
the SonarWatch. We believe that this work can facilitate the skinsensing
technologies in our daily life.
References
HARRISON, C., TAN, D., AND MORRIS, D. Skinput: appropriating
the body as an input surface. In Proc. ACM CHI ’10, 453–
462.
Figure 3: Gestures supported by SonarWatch. (a) and (b) are Click
gesture. (c) and (d) are Slide-to-Select gesture. (c)and (e) are Slide-
Up gesture.
3 Application
We implement three different applications for demonstrating the effectiveness
of our approaches: photo browsing, playlist controlling,
and augmenting Kinect for gaming. The first two applications are
implemented by ActionScript 3.0 and Flash, and the third application
is implemented by OpenCV and OpenNI2.
3.1 Photo browsing
Rich gestures can help user browse photo more efficiently. In
our photo browsing application, user can Slide-Up/Down for turning
pages up/down, Slide-to-Select thumbnail for browsing photos,
Single-click to enter selected album, and Double-click to return.
3.2 Playlist controlling
User may want to control background application such as their
music player in an eyes-free manner. In playlist controlling application,
user can Single-click to play/pause the music, Slide-to-
Select to adjust volume, and Slide-Up/Down for switching to the
next/previous song.
3.3 Augmenting Kinect for gaming
Utilizing body as controller may somehow lack-of-haptic sensation.
Hence, our proprioception can be utilized as complement. We use
the Angry Birds3 game for illustration. By using Kinect to capture
user’s arm position, user can adjust the angle of bow. For launching,
user can Slide-to-Select the desired intensity, then remove their
finger to launch the Angry Birds.
2http://www.openni.org
3http://www.angrybirds.com/
4 Conclusion
We present SonarWatch, By equipping sensor on the readily available click and To demonstrate this approach, controlling and the SonarWatch. technologies References
HARRISON, C., the body 462.
Figure 3: Gestures supported by SonarWatch. (a) and (b) are Click
gesture. (c) and (d) are Slide-to-Select gesture. (c)and (e) are Slide-
Up gesture.
3 Application
We implement three different applications for demonstrating the effectiveness
of our approaches: photo browsing, playlist controlling,
and augmenting Kinect for gaming. The first two applications are
implemented by ActionScript 3.0 and Flash, and the third application
is implemented by OpenCV and OpenNI2.
3.1 Photo browsing
Rich gestures can help user browse photo more efficiently. In
our photo browsing application, user can Slide-Up/Down for turning
pages up/down, Slide-to-Select thumbnail for browsing photos,
Single-click to enter selected album, and Double-click to return.
3.2 Playlist controlling
User may want to control background application such as their
music player in an eyes-free manner. In playlist controlling application,
user can Single-click to play/pause the music, Slide-to-
Select to adjust volume, and Slide-Up/Down for switching to the
next/previous song.
3.3 Augmenting Kinect for gaming
Utilizing body as controller may somehow lack-of-haptic sensation.
Hence, our proprioception can be utilized as complement. We use
the Angry Birds3 game for illustration. By using Kinect to capture
user’s arm position, user can adjust the angle of bow. For launching,
user can Slide-to-Select the desired intensity, then remove their
finger to launch the Angry Birds.
2http://www.openni.org
3http://www.angrybirds.com/
4 Conclusion
We present SonarWatch, By equipping sensor on the wristwatch, readily available click and slide To demonstrate this approach, three controlling and augmenting the SonarWatch. technologies References
HARRISON, C., the body as 462.
Figure 3: Gestures supported by SonarWatch. (a) and (b) are Click
gesture. (c) and (d) are Slide-to-Select gesture. (c)and (e) are Slide-
Up gesture.


3 Application
We implement three different applications for demonstrating the effectiveness
of our approaches: photo browsing, playlist controlling,
and augmenting Kinect for gaming. The first two applications are
implemented by ActionScript 3.0 and Flash, and the third application
is implemented by OpenCV and OpenNI2.
3.1 Photo browsing
Rich gestures can help user browse photo more efficiently. In
our photo browsing application, user can Slide-Up/Down for turning
pages up/down, Slide-to-Select thumbnail for browsing photos,
Single-click to enter selected album, and Double-click to return.
3.2 Playlist controlling
User may want to control background application such as their
music player in an eyes-free manner. In playlist controlling application,
user can Single-click to play/pause the music, Slide-to-
Select to adjust volume, and Slide-Up/Down for switching to the
next/previous song.
3.3 Augmenting Kinect for gaming
Utilizing body as controller may somehow lack-of-haptic sensation.
Hence, our proprioception can be utilized as complement. We use
the Angry Birds3 game for illustration. By using Kinect to capture
user’s arm position, user can adjust the angle of bow. For launching,
user can Slide-to-Select the desired intensity, then remove their
finger to launch the Angry Birds.
2http://www.openni.org
3http://www.angrybirds.com/
4 Conclusion
We present SonarWatch, a less-obtrusive, wearable By equipping an ultrasonic rangefinder and a sensor on the wristwatch, user’s forearm can be appropriated readily available input surface for interacting with click and slide gestures are allowed to be performed To demonstrate the possible applications and the this approach, three applications: photo browsing, controlling and augmenting Kinect for gaming, are the SonarWatch. We believe that this work can facilitate technologies in our daily life.
References
HARRISON, C., TAN, D., AND MORRIS, D. Skinput: the body as an input surface. In Proc. ACM 462.
Figure 3: Gestures supported by SonarWatch. (a) and (b) are Click
gesture. (c) and (d) are Slide-to-Select gesture. (c)and (e) are Slide-
Up gesture.
3 Application
We implement three different applications for demonstrating the effectiveness
of our approaches: photo browsing, playlist controlling,
and augmenting Kinect for gaming. The first two applications are
implemented by ActionScript 3.0 and Flash, and the third application
is implemented by OpenCV and OpenNI2.
3.1 Photo browsing
Rich gestures can help user browse photo more efficiently. In
our photo browsing application, user can Slide-Up/Down for turning
pages up/down, Slide-to-Select thumbnail for browsing photos,
Single-click to enter selected album, and Double-click to return.
3.2 Playlist controlling
User may want to control background application such as their
music player in an eyes-free manner. In playlist controlling application,
user can Single-click to play/pause the music, Slide-to-
Select to adjust volume, and Slide-Up/Down for switching to the
next/previous song.
3.3 Augmenting Kinect for gaming
Utilizing body as controller may somehow lack-of-haptic sensation.
Hence, our proprioception can be utilized as complement. We use
the Angry Birds3 game for illustration. By using Kinect to capture
user’s arm position, user can adjust the angle of bow. For launching,
user can Slide-to-Select the desired intensity, then remove their
finger to launch the Angry Birds.
2http://www.openni.org
3http://www.angrybirds.com/
4 Conclusion
We present SonarWatch, By equipping sensor on the wristwatch, readily available click and slide To demonstrate this approach, three controlling and the SonarWatch. technologies References
HARRISON, C., the body 462.
(a)
(b) (c)
(d) (e)
11年10月14日星期五
Figure 2: Available gestures. (a) Posture for a click. (b) Posture
for a slide. (c) Slide-to-Select. (d) Slide-Up. (e) Slide-Down.
3 Applications
We have implemented three different applications for demonstrating
the effectiveness of our approaches: photo browsing, playlist
controlling, and augmenting Kinect for gaming.
Photo browsing: Gestures can help users browsing photos more
efficiently. In the photo browser, the users can Slide-Up/Down
for turning pages, Slide-to-Select thumbnails for browsing photos,
Single-click to select an album, and Double-click to return.
Playlist controlling: While working with listening to the music,
users may want to control the playlist in an eyes-free manner.
Hence, the users can Single-Click to play/pause, Slide-to-Select to
adjust volume, and Slide-Up/Down to switch to another song.
Augmenting Kinect for gaming: Utilizing body as controller may
somehow lacking haptic sensation, but we can use our proprioception
as a complement. We illustrate this idea by the Angry Birds2
game. Since a user’s arm position can be captured by Kinect, she
can perform a shot by moving her arm up and down to adjust the angle
of slingshot, sliding the fingers on the arm to select the desired
intensity, and removing the fingers to launch the Angry Birds.
References
HARRISON, C., TAN, D., AND MORRIS, D. 2010. Skinput: appropriating
the body as an input surface. In Proc. ACM CHI ’10,
453–462.
NAKATSUMA, K., SHINODA, H., MAKINO, Y., SATO, K., AND
MAENO, T. 2011. Touch interface on back of the hand. In ACM
SIGGRAPH 2011 E-Tech., 39.
2http://www.angrybirds.com/

晃晃

谢谢楼主，真是太实用了

奇

我就看看，我不说话

奇

很经典，很实用，学习了!

C.R.CAN

俺是新人,这厢有礼了!

菜刀吻电线

非常感谢，管理员设置了需要对新回复进行审核，您的帖子通过审核后将被显示出来，现在将转入主题

奇

不错哦,顶一下......

tc

真不错,全存下来了.

tc

好铁多多发，感激分享