MSc-thesis Ideas for Chalmers-NUS collaborations Rationale: Next are some ideas for the Chalmers-NUS collaboration. The plan is the Chalmers students pick one of these themes, elaborate, and suggest a detailed Master thesis plan. This detailed plan should fulfill the criteria for a thesis proposal, given at: http://ixdcth.se/courses/2012/ciux02/ and https://chalmersuniversity.box.com/shared/static/zs253mogen3j6weeu59k.pdf The process will follow the steps proposed in a separate timeline document.

 

Theme 1A: Color Blind: Haptic Glove PIs NUS: PIs Chalmers:

Shen Paweł

Description: While color-blindness is not a life-threatening illness, it does produce difficulties in everyday life. This project investigates how technology can help color blind in performing daily lives tasks, such as shopping for cloths. The project aims at the development of a haptic glove that is capable of helping color blinds in their daily lives. The current prototype is able to detect and present colors to the user and more work is needed to user test the prototype and propose improvements based on users’ feedback. The focus of the work will be on integrating technology into everyday routines through field studies. Required background:

Electronic prototyping, programming, system integration, and user study

Project-URL: Potential conference(s):

AH 2015, submission 15th November 2014

Theme 1B: Color Blind: Hands Free/Eyes Interaction PIs NUS: PIs Chalmers:

Shen Morten

Description: The most initiative way for color blinds may well be a hands-free way to detect colors. This means, we do not need to use a glove, but maybe detect colors based on where the user is looking. This can be done with the help of an eye-gaze tracker and with a device such as Google glass. Required background:

Electronic prototyping, programming, system integration, and user study

Project-URL:

http://www.shengdongzhao.com/publication/exploring-user-motivationsfor-eyes-free-interaction-on-mobile-devices/

 

Theme 2: Social Robot: Cross-Cultural Study PIs NUS: PIs: Chalmers:

Shen Mohammad

Description: One’s personal social status is shaped based on their own achievements and their inherited features. In the field of Sociology, this is called achieved status and ascribed status, respectively. Achieved status is based on the skills and abilities that an individual acquires, such as their education and work, while ascribed status is based on features they are born with, such as gender, race, and family wealth. In this project, we aim at exploring the Human expectation of robots on the bases of social status. The main question here is: Do we care about the social status of robots? Is the answer to this question different for different cultures? Required background:

Robotics, programming, system design, and user studies

Project-URL:

http://www.shengdongzhao.com/shen-publications/#HRI

Technology:

Robot platform, for instance http://www.doublerobotics.com

Potential conference(s):

HRI 2016, submission early October 2015

Theme 3: Crowdsourcing PIs NUS: PIs Chalmers:

Shen and Chen Chen Pawel, Mohammad, and Morten

Background: Bigham et al. [3] state that intelligent systems will be developed more quickly by first creating crowd-powered systems, before gradually moving to fully automated solutions. Accordingly, pushing the scope of problems solvable with crowdsourcing implies a) engaging with expert crowds, b) embedding needed expertise in the tools non-expert crowds use, and c) using a flexible combination of the two approaches. Inspired by previous work [5, 9] we know crowdsourcing can be used to help analyze large amounts of scientific data instead of assigning these time-consuming tasks to experts with higher costs and less availability. This kind of participatory science is a promising new application domain for crowdsourcing. A notable example of this is the foldit project (http://fold.it/portal/info/about), which has developed a game wherein nonexperts help scientists figure out how proteins fold, which is important for understanding how to target them with pharmaceuticals. Indeed, Eiben et al. [6] and several others have successfully carried out and published research using foldit. In a first concrete research effort, we set out to gain increased understanding and helping to prevent disasters in silo storage and transport.  

Work carried out so far in this project: We built a crowdsourcing system designed to perform the detection of tracer particles in tomographic images, applied to the problem of bulk solid flow in silos (figure above). As imagery from silo sensing systems cannot be properly analyzed by computational methods, it requires human intelligence. However, limited availability of experts, as well as their high cost, motivates employing additional non-experts. In this first study, we could assess the precision and performance of employing non-expert users to process large datasets from rich images in order to generate actionable data. Our in-house design system can help analyze and understand the physics of flow and inform the future design of silos to avoid the current problem. In the next steps of this project, we propose three/four topics of investigation in the area of a gaming framework for crowdsourcing work. 1. 2. 3. 4.

A full review of all crowdsourcing methods and applications that exist in literature A full review of all serious game methods and applications that exist in literature. Mapping game methods to crowdsourcing methods, Defining a framework that can be suitable for crowdsourcing applications.

Project-URL:

To be defined (work currently in review)

Potential journal outlet:

ACM Transactions on Intelligent Systems and Technology Special Issue on Crowd in Intelligent Systems Submission due: First review completed: Final manuscript due: Publication:

16 January 2015 17 April 2015 19 June 2015 September 2015

IUI 2016, submission mid-September 2015

Theme 4A: Vignette: Disney Animation Rules PI NUS: PI Chalmers:

Shen Mohammad

Description:

The aim of this project is to support the build a tool with extra parameters that can help in generating animations based on some of the Disney animation rules.

Required background:

Good graphic design skills, simulation-engine, and game design

Project-URL:

Systems and Tools for Arts and Creativity http://www.shengdongzhao.com/shen-publications/#ArtsCreativity AND http://www.shengdongzhao.com/publication/vignette-interactive-texturedesign-and-manipulation-with-freeform-gestures-for-pen-and-inkillustration/

Related work:

http://en.wikipedia.org/wiki/12_basic_principles_of_animation

Potential conference(s):

SIGRAPH ASIA 2015 NPAR 2015, http://expressive2014.mpi-inf.mpg.de/NPAR/Home ACE 2015, http://www.ace2014.info/ACE_2014/Participation.html

Theme 4B: Educational illustrations and animation principles PI NUS: PI Chalmers:

Shen Mohammad and Morten

Description:

The aim of this project is to investigate if animation principles can support educational illustrations (such as mathematical illustrations, physics or chemistry). One part is to explore Disney's Twelve Basic Principles of Animation and define which of the principles can proof to support educational illustrations and understanding

Required background:

Computer graphics, programming, and user studies

Project-URL:

Systems and Tools for Arts and Creativity http://www.shengdongzhao.com/shen-publications/#ArtsCreativity AND http://www.shengdongzhao.com/publication/sandcanvas/

Related work:

http://en.wikipedia.org/wiki/12_basic_principles_of_animation

Video: Potential conferences:

 

SIGRAPH ASIA 2015, submission end May (or early June) 2015 NPAR 2015, http://expressive2014.mpi-inf.mpg.de/NPAR/Home ACE 2015, http://www.ace2014.info/ACE_2014/Participation.html