1. APA Citation :
Chris Harrison, Desney Tan, and Dan Morris. 2010. Skinput: appropriating the body as an input surface. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’10). ACM, New York, NY, USA, 453-462. DOI=10.1145/1753326.1753394 http://doi.acm.org/10.1145/1753326.1753394
2. Purpose :
In this paper, the Authors present a technology called as ‘ Skinput’ – a device that makes the whole body as an input device by analyzing the mechanical vibrations that propagate through the body by touching a part of body skin (here Skin in the Arm region) called Bi-Acoustic Signal Acquisition Process. This device and research paper is another great valuable addition to Wearable Computing and Finger Input Technologies. Also, this device will open doors for an always available ‘System’ with which we can interact anytime.
3. Methods :
i. Construction of the Device :
The Research Scholars built a custom device with the following important components for developing the ‘Skinput’ system. And, they are :
a. Pico-Projector – For projecting the output from the system to the User for Visualization.
b. Bio-Acoustic Sensors – For detecting the User Touches.
c. Arm Band – The Arm band collectively houses the Bio-Acoustic Sensors for the Users to wear them and the arm band itself serves as a place holder for co-ordination of the sensors with that of the arm.
d. Sample Applications – They also built several sample applications which served for testing the Device, User Study and Performance Analysis of the Device.
e. Audio Output Device – For audio feedback.
For clarification, the following image shows the Input Sets and proximity positions of the Bio-Acoustic sensors that should be present in the arm for detection of the user inputs by the device.
ii. Evaluation of the Device :
For evaluating the device, an User Study was conducted with 6 Male and 7 Female participants of an age group ranging from 20 to 56 with different Body Mass Indexes. (The BMI of a person is an important criteria in this evaluation because it directly affects the Vibration Level of the Skin in turn affecting the detection of ‘User Touches’ by the Sensor.)
Apart from the above said testing locations of the skin (i.e., Fingers, Whole Arm and Fore Arm), there was also another set of supplemental testing with respect to the environmental conditions so as to test the device in real-time environment and targeted setting. The following are the supplemental methods used for testing the device :
a. Walking / Jogging
b. Single Handed Gestures
c. Surface and Object Recognition
d. Identification of Finger Tap Type
e. Segmenting Finger Input
The participants were 1 Male and 1 Female for the first exercise. For the rest of the experiments, 7 new participants (3 Females) were invited for the testing the device.
4. Main Findings :
After the User Study, the following were the main results :
a. The device performed well with an average accuracy of 87.6 % from the all experiments.
b. Additionally, the results of the experiments were also analyzed by considering the Body Mass Index, Sex and Age of the Participants for improving the device in the future.
c. By numbers, the device was very much and well capable in detecting the inputs from the Male participants in higher rate than that of female participants.
d. From the overall results of the supplemental tests performed, the environmental conditions and targeted test settings did not affect the performance of the device.
5. Analysis :
In every front, this was a very good research. It has the potential to be successful if it is harnessed and used in the right way, so as it is to be usable and useful in every dimension. The authors included several methods for testing the device which is very much convincing for me that the user study is performed well. Also, separated indexing of the impact of BMI analysis in the device was good.
Personally, though I cannot find any flaws in this paper, I felt that this paper was not much of a phenomenal game changer but an incremental research. I chose this paper as it was kind of similar to one of my favorite research articles of all time – SixthSense by the MIT Media Lab. This paper being just two years old has more than 96 citation in total. I am interested in following up this paper’s citations and the future research direction of the research scholars.