Marlon Richert, Master’s <strong>The</strong>sis, Friday, 23 February, 2007 vi
1 Introduction Marlon Richert, Master’s <strong>The</strong>sis, Friday, 23 February, 2007 <strong>The</strong> Medical Imag<strong>in</strong>g & <strong>Co</strong>mput<strong>in</strong>g Laboratory (MedICLab) at the Center for Research and Innovation <strong>in</strong> Bioeng<strong>in</strong>eer<strong>in</strong>g (Ci 2 B or Centro de Investigación e Innovación en Bio<strong>in</strong>geniería) of the Universidad Politécnica de Valencia (UPV) <strong>in</strong> Spa<strong>in</strong> has developed and is develop<strong>in</strong>g several virtual therapy tools—virtual reality (VR) tools for use <strong>in</strong> various k<strong>in</strong>ds of therapy. One such tool is the Book of Life, <strong>in</strong>tended for help<strong>in</strong>g patients deal<strong>in</strong>g with traumatiz<strong>in</strong>g experiences by lett<strong>in</strong>g them create a visual story out of key periods of their life us<strong>in</strong>g symbolic elements, under the guidance of a therapist. Another example is the Virtual Reality <strong>Mirror</strong>, a system for stimulat<strong>in</strong>g bra<strong>in</strong> plasticity <strong>in</strong> hemiplegics, while m<strong>in</strong>imiz<strong>in</strong>g the need for therapeutic guidance. Ideas for other virtual therapy tools are underway. One such idea, for example, is a food therapy tool <strong>in</strong> which the patient virtually prepares and eats food, <strong>in</strong> order to prepare them for do<strong>in</strong>g so <strong>in</strong> the real world. Right now, most of the MedICLab’s virtual reality tools use a head-mounted display (HMD) for stereoscopic view<strong>in</strong>g and a gamepad for user <strong>in</strong>put. One problem with these systems is HMDs are quite <strong>in</strong>vasive and can be uncomfortable to use, while a gamepad is not a very <strong>in</strong>tuitive way of control for most users, especially those unfamiliar with them. <strong>The</strong> Virtual Reality <strong>Mirror</strong> is an exception to this as it does not require an HMD and users motion trackers <strong>in</strong>stead of a gamepad, but this system does not provide stereoscopy and only limited immersion. <strong>The</strong> MedICLab develops all of its VR applications—<strong>in</strong>clud<strong>in</strong>g systems other than virtual therapy tools, such as surgical simulations and computer-assisted medical <strong>in</strong>tervention tools—us<strong>in</strong>g a proprietary third-party development tool called IPF, orig<strong>in</strong>ally <strong>in</strong>tended for creat<strong>in</strong>g virtual TV studio sets. <strong>The</strong> MedICLab has <strong>in</strong>vested a considerable amount of f<strong>in</strong>ances and human resources <strong>in</strong> IPF. Unfortunately, much of IPF’s functionality uses a non-standard paradigm and has only poor documentation, mak<strong>in</strong>g it hard for newcomers to MedICLab to master work<strong>in</strong>g with the tool. Additionally, although IPF is very suitable for creat<strong>in</strong>g static virtual environments, because of its orig<strong>in</strong>s, it has only very poor support for dynamic <strong>in</strong>teractions. Instead, its ma<strong>in</strong> <strong>in</strong>teraction focus lies mostly <strong>in</strong> manag<strong>in</strong>g simple buttons and triggers—such as those required for live TV coverage of national elections, for example—not the k<strong>in</strong>d of complex <strong>in</strong>teractions required by direct manipulation <strong>in</strong> virtual reality. Thus, creat<strong>in</strong>g these <strong>in</strong>teractions often proves complex and time-consum<strong>in</strong>g. F<strong>in</strong>ally, the MedICLab would like to move most of their virtual reality projects and ideas to augmented reality (<strong>AR</strong>). One reason for this is that they want to explore whether it <strong>in</strong>creases immersion, which is important for many of their tools. Another reason is that, with <strong>AR</strong>, the patient can still observe the real world, <strong>in</strong>clud<strong>in</strong>g the therapist, mak<strong>in</strong>g it a more comfortable and less alien experience. <strong>The</strong> MedICLab evaluated several <strong>AR</strong> approaches and decided that their preferred choice of <strong>AR</strong> would be mirror-based see-through <strong>AR</strong>. <strong>The</strong> ma<strong>in</strong> reasons for this are that it is non<strong>in</strong>vasive and easily allows for co<strong>located</strong> reach-<strong>in</strong> <strong>in</strong>teraction. Additionally, it has the potential to be relatively <strong>in</strong>expensive, which is a requirement because they do not have a large budget. F<strong>in</strong>ally, this k<strong>in</strong>d of solution is the most suitable situation for most of their projects, with the Virtual Reality <strong>Mirror</strong> <strong>in</strong> particular. Thus, the research question of this thesis is as follows: Is it possible to develop an <strong>in</strong>expensive yet complete co<strong>located</strong> reach-<strong>in</strong> mirror-based see-through augmented reality solution that is non<strong>in</strong>vasive and <strong>in</strong>tuitive to the user, but requires m<strong>in</strong>imal doma<strong>in</strong> knowledge from the application programmer and is not tied to a particular 1
- Page 1 and 2: THE AR WORKBENCH: A COMPLETE COLOCA
- Page 3: PREFACE Marlon Richert, Master’s
- Page 8 and 9: Marlon Richert, Master’s Thesis,
- Page 10 and 11: Marlon Richert, Master’s Thesis,
- Page 12 and 13: Marlon Richert, Master’s Thesis,
- Page 14 and 15: Marlon Richert, Master’s Thesis,
- Page 16 and 17: Marlon Richert, Master’s Thesis,
- Page 18 and 19: Marlon Richert, Master’s Thesis,
- Page 20 and 21: Marlon Richert, Master’s Thesis,
- Page 22 and 23: Marlon Richert, Master’s Thesis,
- Page 24 and 25: Marlon Richert, Master’s Thesis,
- Page 26 and 27: Marlon Richert, Master’s Thesis,
- Page 28 and 29: Marlon Richert, Master’s Thesis,
- Page 30 and 31: Marlon Richert, Master’s Thesis,
- Page 32 and 33: Marlon Richert, Master’s Thesis,
- Page 35 and 36: 6 Implementation Marlon Richert, Ma
- Page 37 and 38: Marlon Richert, Master’s Thesis,
- Page 39 and 40: Marlon Richert, Master’s Thesis,
- Page 41 and 42: Marlon Richert, Master’s Thesis,
- Page 43 and 44: Marlon Richert, Master’s Thesis,
- Page 45: Marlon Richert, Master’s Thesis,
- Page 48 and 49: Marlon Richert, Master’s Thesis,
- Page 50 and 51: Marlon Richert, Master’s Thesis,
- Page 53 and 54: REFERENCES Marlon Richert, Master
- Page 55: Marlon Richert, Master’s Thesis,
- Page 58 and 59:
Marlon Richert, Master’s Thesis,
- Page 60 and 61:
Marlon Richert, Master’s Thesis,
- Page 62 and 63:
Marlon Richert, Master’s Thesis,