Group Number 16: Epple
Andrew, Brian, Kevin, and Saswathi
Summary:
Our project is an interface through which controlling web cameras can be as intuitive as turning one’s head.
Introduction:
Our system uses a Kinect to monitor a person’s head orientation and uses this to remotely control the angle of a servo motor on top of which a web camera is mounted. This essentially allows a user to remotely control the camera view through simple head movements. The purpose of our system is to enable the user to intuitively and remotely control web cameras and thus engage in a realm of new web chat possibilities. Normally web chat sessions end up being one-on-one experiences that fall apart once a chat partner leaves the view of the camera. With our system, we aim to allow for more dynamic web chat sessions in which there may be multiple chat partners on the other side of the camera, and these partners can move freely.
Implementation and Improvements:
Our P5 submission is located at: http://blogs.princeton.edu/humancomputerinterface/2013/04/22/p5-group-16-epple/
Since P5, we have not implemented any changes to our working prototype.
Method:
Participants:
Participants were selected from students in the Frist campus center. They were chosen semi-randomly out of the students who were looking for a study break and agreed to help up test our product. Our first participant was Chaithanya Yarlagadda who is a Fulbright scholar from India and is a PhD student in Wildlife Biology who one of our team members had briefly met earlier in the year and whom we asked because we felt he would have a unique perspective, as he is an international student, older than the average undergraduate, and video chats fairly often. Our second participant was Enyonam Glover who is a senior undergraduate in the sociology department. She was getting a meal at the Frist student center, and we asked her to test our prototype as she ate her meal, as many people often video chat while eating. She also frequently video chats with her family, about twice a week. Our third participant was Kevin Ofori who is a senior undergrad in the Economics department. We discovered later, actually, that he is opposed to using web chat on principle, but we invited him to test our prototype since he was passing by and we felt that he could offer a unique point of view.
Apparatus:
Equipment that we used strictly for our hi-fidelity prototype includes a Laptop, a webcamera, an arduino, a kinect and an ipad. In addition, we also had to situate the prototype so that it simulates distance between the user and the chat partner because we have not yet implemented networking functionality. Because of this, the Kinect we used to detect user head movements was connected to the same computer as the remote, rotating webcamera. We thus had to select a location where the user could sit within view of the kinect while having the remote camera view a different scene from what was directly in front of the user. The bend of a corner served this purpose nicely, as we were able to simply put the user and the Kinect on one side of the corner and the remote web camera on the other side of the corner, all while connecting both components to the same computer. We chose to conduct our tests at Frist Campus Center because of its central location, as we would be able to easily find volunteers for user testing. Some of our tasks also require high person-density rooms, a role that Frist Campus center wholly fulfills.
Tasks:
We have not changed tasks from P5. The first task we have chosen to support with our working prototype is the easy-difficulty task of allowing for web chat while breaking the restriction of having the chat partner sit in front of the computer. A constant problem with web chats is the restriction that users must sit in front of the web camera to carry on the conversation; otherwise, the problem of off-screen speakers arises. With our prototype, If a chat partner moves out of the screen, the user can eliminate the problem of off-screen speakers through intuitive head movements to change the camera view. Our second task is the medium-difficulty task of searching a distant location for a person with a web camera. Our prototype allows users to seek out people in public spaces through using intuitive head motions to control the search of a space via web camera, just as they would in person. Our third task is the hard-difficulty task of allowing web chat with more than one person on the other side of the web camera. Our prototype allows users to web chat seamlessly with all the partners at once. Whenever the user wants to address a different web chat partner, he will intuitively change the camera view with his head to face the target partner.
Procedure:
We went to Frist Campus Center and found our volunteers. Once found, we made a volunteer read and sign a consent form. We also asked them to fill in a demographic information questionnaire. We then presented our system using a demo and task script that we already wrote where we presented our tasks and demo’ed the system. We also made sure to warn them about limitations of our system such as the Kinect being unable to track the user’s face behind solid objects or at extreme angles. We then allowed our users to perform all three tasks while noting when errors occurred and Task 2 completion time. Finally, we asked the users to fill in a post-task questionnaire and give us general feedback on our system.
Test Measures:
As our main goal is to achieve an intuitive interface, we measured self-reported measures of satisfaction through asking our participants to fill in a post-task questionnaire. This questionnaire had several statements that the participant rated using a standard Lickert Scale. We felt this self-report was the most suitable form of measurement for intuitiveness, as intuitiveness is not something that is easily observable. Rather, intuitiveness is a fairly abstract, subjective concept, and thus should be measured through self-reported means. From these self-reports we measured:
- If the interface was intuitive
- If the interface was tiring to use
- If the interface allowed for easily doing things that were not possible before
- If the interface was something that the user would often use
- If the concept of the interface was good
- If the execution of the interface was good
- If the task 1 was easy to do
- If the task 1 was useful
- If the task 2 was easy to do
- If the task 2 was useful
- If the task 3 was easy to do
- If the task 3 was useful
Other, more observable, means of measuring intuitiveness that we also used include:
- Task 2 completion time
- Number of errors
Number of errors is a fairly nice way to measure intuitiveness, as the more intuitive the design is, the less error-prone users should tend to be. From the same logic, we also reason that task 2 completion times should be lower when more intuitive interfaces are used.
Results and Discussion:
From our measurements of number of errors, we saw that there is a small initial learning curve associated with using our interface. When a person first uses the interface, they tend to make a few errors in the first seconds. During this period, they appear to be calibrating their minds with our system and making associations between actions and outcomes. We reason that this learning period naturally exists because our interface is extremely unique. For many people, our interface was their first experience of remote controlling an object through their body movements. After the first few seconds, we found that users quickly adjusted and then proceeded through the tasks naturally with zero further errors. From this, we conclude that our system is indeed intuitive to use as it is a simple one-to-one mapping of body movement to camera movement.
We found that all task 2 completion times were on the order of 2 to 3 seconds which is extremely fast. During the actual testing process, we observed that users accomplished the tasks as quickly and naturally as if they were in the remote area in person. We also found from the questionnaire results that our participants all agreed that the tasks they were asked to do were easy to do and useful. In fact, all of our participants either “agreed” or “strongly agreed” with statements saying that the tasks were easy to do. Also, none of our participants “disagreed” or “strongly disagreed” with statements saying that the tasks were useful. From these two sets of measurements, we conclude that we successfully created an interface through which it is easy to accomplish the three useful tasks.
Based on the questionnaire results, while there was agreement that our tasks were easy to do and useful, there was dissenting opinion over how useful our system was as a whole. Participant 1, in particular, was neutral when discussing its potential usefulness since he found current systems already adequate and had no need for a moving camera in web chats. He did say, however, that there might be situations (if a parent wanted to play a game with his/her child via web chat) where it might have a niche use. Participant 2 was very positive, saying that she would love to see it in production. Participant 3, on the other hand, strongly disagreed that he would use it often, citing a fear that it could be abused, militarized, and that he had a current policy of avoiding using web chat (saying that they took too much time and invaded his privacy). Due to the disagreement among our members and the confounding influences behind participant 3’s rating, we would be hesitant to draw any firm conclusions on the usefulness of our system as a whole based on the results of our study. Actually, whether something is useful or not may be a matter that is entirely too subjective to analyze quantitatively from just three participants. At the very least, we believe that we need a much larger sample size to test before we can attempt to draw any definitive conclusions on this matter.
Also based on the questionnaire results, we find that most of our participants agreed that our system, as a whole, was intuitive. Kevin Ofori, however, “strongly disagreed” with the statement that our interface was intuitive. This was not to say that our system was hard to use, as he too agreed with the tasks being easy to do. Rather, he wanted to express fear that our system may, in the future, become irrelevant when Google Glass is released. He mentioned that it was hindering and unnatural to always have to hold the mobile iPad screen in front of himself in order to web chat. He would rather have the mobile screen automatically move to be in front of his head as he moved. We reason, however, that once Google Glass is released, rather than making our system irrelevant, we can simply incorporate Google Glass into our system. After all, the main functionality of our system is to remotely control a web camera through head movements; this is something that Google Glass cannot do by itself. Thus, the actual mobile screen component of our system is something that can easily be replaced. Once wearable screens such as Google Glass and Oculus Rift come out, we can simply replace our mobile iPad component of the interface with these new technologies to further enhance the intuitiveness. In the meantime, we plan to rectify this problem with a change to our prototype. We can solve the problem by attaching the iPad to the arm of a swiveling chair so as to fix the iPad in front of the user. The user can then just rotate in the chair to change the camera view while the iPad is automatically kept in front of him.
Appendices:
Consent Form (same as previous)
[LINK]
Demographic Questionnaire:
What is your age?
What is your sex?
What is your education level?
What is your major?
What is your experience with web chats?
Demo Script:
Hello, you have been invited to test our prototype for an interface for web camera control. The purpose of our interface is to allow a user to intuitively control a web camera through simple head movements that will be viewed by a Kinect. You will be given an iPad with the Skype app running with which you can view the feed of a web camera. You can naturally change your head orientation, and the camera view will change correspondingly. Here we will demo one task so that you can better understand our system. This is Saswathi; we’re having a skype conversation with her right now. Our interface allows us to change the angle of the mounted webcam in exactly the same way we move our head. [Demonstrate, being careful to swivel our arms in tandem with our head]. Do note, however, that the Kinect is unable to track your head if it is turned to an extreme angle. [Demonstrate]. Also, it requires full view of your face to track it, so please be careful not to hold the iPad too high or close to your face.
Task Script 1: Saswathi moves around
The first task that we want you to do is to web chat while breaking the restriction of having your chat partner sit in front of the computer. With a typical interface, this scenario would just cause your partner to go off screen, but with our interface, you can now simply move your head to look and talk to a target person as he moves. In the task, the person may move around the room, but you must keep the target within view.
Task Script 2: Find Waldo
The second task is to be able to search a distant location for a person through a web camera. While you might seek out a friend in Frist to initiate a conversation, in web chat, the best you can do is wait for said friend to get online. We intend to rectify this by allowing users to seek out friends in public spaces by searching with the camera, just as they would in person. You will play the “Where’s Waldo” game with Saswathi. There are various people in Frist, and you will need to use our interface to move a camera view around until you are able to find Saswathi.
Task Script 3: Family Chat
The third task is to web chat with more than one person on the other side of the web camera. A commonly observed problem with web chats is that even if there are multiple people on the other end of the web chat, it is often limited to being a one on one experience where chat partners wait for their turn to be in front of the web camera. We will have multiple people carrying a conversation with you, and you will be able to view the speakers only through the screen. You can turn your head in order to address a particular conversation partner. When you hear an off-screen speaker, you may then turn your head to focus on him.
Post-task Questionnaire:
Please indicate whether you strongly agree, agree, disagree or strongly disagree with, or are neutral about the following statements:
– This interface was intuitive.
– This interface was tiring to use.
– This interface allows me to easily do things that were not possible before.
– This is something I would use very often.
– This concept is good.
– This execution is good.
– Task 1 was easy to do.
– Task 1 is useful.
– Task 2 was easy to do.
– Task 2 is useful.
– Task 3 was easy to do.
– Task 3 is useful.
Raw Data:
Participant 1:Chaithanya Yarlagadda
Age: 33
Sex: Male
Education level: PhD Student
Major: wildlife biology
Experience: ~webchat 3 times a week with family and friends
Lickert Scale:
This interface was intuitive.
Agree
This interface was tiring to use.
Neutral
This interface allows me to easily do things that were not possible before.
Neutral
This is something I would use very often.
Disagree
This concept is good.
Neutral
This execution is good.
Agree
Task 1 was easy to do.
Agree
Task 1 is useful.
Neutral
Task 2 was easy to do.
Agree
Task 2 is useful.
Agree
Task 3 was easy to do.
Neutral
Task 3 is useful.
Neutral
Notes:
4 errors in first few seconds of use
Task 2 completion time: 2 sec
general comments:
– most normal situations fine with a stationary camera.
– if you want to play a game with your kid this would be useful
Participant 2: Enyonum Glover
Age: 21
Sex: Female
Education level: Undergrad
Major: Sociology
Experience: 2x per week
Lickert Scale:
This interface was intuitive.
Agree
This interface was tiring to use.
Disagree
This interface allows me to easily do things that were not possible before.
Strongly agree
This is something I would use very often.
Strongly agree
This concept is good.
Strongly agree
This execution is good.
Neutral
Task 1 was easy to do.
Agree
Task 1 is useful.
Strongly agree
Task 2 was easy to do.
Strongly agree
Task 2 is useful.
Strongly agree
Task 3 was easy to do.
Agree
Task 3 is useful.
Strongly agree
Notes:
1 error in first few seconds of use
Task 2 completion time: 2 sec
general comments:
– hard to understand setup of system because there are many parts
– easy to use once she got used to it
Participant 3: Kevin Ofori
Age: 21
Sex:male
Education:undergrad senior
Major: econ
Experience: try to avoid it – skype once a month
Lickert Scale:
This interface was intuitive.
Strongly disagree
This interface was tiring to use.
Agree
This interface allows me to easily do things that were not possible before.
Disagree
This is something I would use very often.
Strongly disagree
This concept is good.
Agree
This execution is good.
Agree
Task 1 was easy to do.
Strongly agree
Task 1 is useful.
Agree
Task 2 was easy to do.
Strongly agree
Task 2 is useful.
Strongly agree
Task 3 was easy to do.
Agree
Task 3 is useful.
Agree
Notes:
5 errors in first few seconds of use
Task 2 completion time: 3 sec
general comments:
– google glass is cooler but costs more
– the first thing that is gonna happen is it’ll be militarized
– it’ll be an add-on to skype
– it might be an invasion of privacy…but not more then having another person in your room
– if you could connect the ipad to your neck with a brace…it would be nice