Today I attended the graduation ceremony for all 2013 VET Development Centre Specialist Scholars. The graduation ceremony was the final event for all recipients of this year’s Specialist Scholarship. The event featured an opening address and presentation of certificates and a gifts to all scholars from VET Development Centre CEO Denise Stevens.
The graduation ceremony also featured an amazing presentation from Terese McAleese, Director of Learning at Swinburne University of Technology. Terese spoke about her journey as an immigrant from 1980s Belfast to Melbourne as well as her journey in vocational education and training (VET), from a student to an administrator and teacher and finally to a professional responsible for designing learning experiences. Terese also spoke about the transformational nature of the VET sector and TAFE, a place where skills could be developed for the direct application in industry as well as a place for second chances or opportunities for education that might have been missed the first time. The chance for a new or better life.
Placing geometry in the 3D point cloud in Metaio Creator 2.6 is a little cumbersome, bewildering and often inaccurate. It takes lots of clicking and seemingly random fiddling to place geometry. You also can’t really determine if your geometry is placed correctly within the 3D point cloud until you upload to your channel and then view your channel in Junaio on your device. If it’s not placed correctly, you need to tweak it in Creator, publish it and then view it again in Junaio. Repeat process until it looks like it might almost be placed correctly. Curious.
Screen captures from my image tracking tests in Junaio for each step of the disassembly process with UI elements. Unfortunately, due to my Z translation problems in UI Designer of Creator 2.6 I had to remove the background image which was placed underneath the buttons. Doing this gives the buttons a floaty and slightly random feel, but at least the buttons function as they should!
I’m currently wrestling with the correct Z translation of images and buttons in UI Designer of Creator 2.6. I’ve placed some buttons on top of an background image in the UI Designer of Creator and have encountered a problem where the image obscures the buttons and prevents them from working. I’ve checked the Z translation of the image in relation to the buttons and have made sure the buttons translated above the background image, but this doesn’t seem to make a difference when the published channel is viewed in Junaio on my iPad.
While attempting to solve this problem on my own in Creator, I experimented with placing the buttons partially over the background image and then publishing to my channel. This had an interesting result. The button displays and functions correctly!
Here are some screen captures of the properties windows for the background image and each button in the UI Designer of Creator and the published channel viewed in Junaio on my iPad.
Today I attended the third and final scheduled event for the VET Development Centre Specialist Scholarship. The event was a knowledge sharing event where all specialist scholars presented their findings and outcomes from their participation in the program. Unlike Event 1 and Event 2, this event gave everyone the opportunity shared what they’ve learned.
The Tweets were correct! The diversity of presentations from each scholar and the range of application and utilisation was amazing. From study programmes, workshops, conferences to the creation of artifacts, each scholar had made the most of their time in a useful and productive way. Thanks VET Development Centre for giving us all the opportunity to take part in the Specialist Scholar program.
After two days of waiting, my Step 4: Remove piston seal from caliper and Step 1: Remove bracket from caliper channels on Junaio have been made public. This is good news, but I think I went a little early on the publish because I failed to include extras such buttons for the instructional video, learner resources and link to next step in the process. The geometry is also misplaced. Now trying to upload to update the published channel and it doesn’t seem to be working. You might have to unpublish the channel, upload changes, publish it and then wait two more days for it to be approved.
Nope, you can just upload the Creator file to the server again and it will ask you to update the already published channel. For me, Creator and the Metaio/Junaio platform lack some expectation-outcomes-experience-scaffolding for users. This software, platform and service can be a little bewildering at times.
This recording shows my first two channels published to Junaio. Both channels feature incorrect placement of geometry and missing user interface. It’s a work in progress.
Previsualising the pointed tool, piston seal and rear brake caliper geometry for Step 4: Remove piston seal from caliper stage of the brake caliper augment. Geometry will be exported as FBX from Blender, prepared by FBXMeshCoverter for import into Creator for upload into my Metaio channel for final use as an augment. Figuring out the production workflow for each 3D model used in each step of the augmented contextual instruction.
Important tip: Remember to export your fbx from Blender to your Desktop and not anywhere else on your computer. For some unknown or unexplained reason fbx files exported to anywhere else but the Desktop do not seem to work properly when converted using FBXMeshConverter.
Screen captures from 3D point cloud data gathering with Metaio Toolbox for sequences to be used in my augmented contextual instruction on the rear brake caliper. The 3D point cloud data will then be used as tracking technology to place my augments.
Captions can help to provide an equivalent learning experience for viewers who may be hearing impaired, speak other languages or use assistive technology. Captions are also valuable in a teaching and learning context where it may be impractical for learners to wear headphones or play video at high volume in an group training environment such as workshop, classroom or laboratory.
Previsualising piston, boot and arrow geometry for Step 3: Remove piston from caliper stage of the brake caliper augment. Geometry will be exported as FBX from Blender, prepared by FBXMeshCoverter for import into Creator for upload into my Metaio channel for final use as an augment. Figuring out the production workflow for each 3D model used in each step of the augmented contextual instruction.
Preparing low-poly geometry for augmented contextual instruction in the disassembly and assembly of a vehicle’s rear brake caliper. Augmented contextual instruction is to be prepared in Metaio Creator, published to a Metaio channel and then accessed by learners through the Junaio app. 3D point cloud data is gathered with Metaio Toolbox.
This video conceptualises one stage of the disassembly of a brake caliper.
[Edit: Tuesday 13 August 2013] This video incorrectly conceptualises one part of the brake caliper disassembly process. A screwdriver is not used to extract the piston, seal and rubber boot from the caliper.
For this prototype, I’m using Metaio Creator image and object recognition features of Toolbox in restrictive Demo Mode to further explore some aspects of my concept of augmented contextual instruction. Unfortunately, in Demo Mode I can’t use the excellent 3D point cloud data captured in Toolbox to prototype all aspects of my augmented contextual instruction. In Demo Mode augments can only be triggered by a QR code, which is kinda okay for testing while you’re building. I’m thinking about buying a license.
This video shows some of the features of the augmented contextual help I’m trying to prototype with Metaio Creator in Demo Mode.
Augmented contextual instruction provides users with procedural demonstrations based on recognisable features and attributes of an object. Augmented contextual instruction could be used in vocational training and assessment contexts. Users can add (record), edit and share their own contextual instruction.
While not specifically working in the filed of AR, Thomas Tucker from Winston-Salem State University spoke about his work with high-end technology such as the FARO scanner and consumer-level technology such as the Kinect. We also got to play with FARO scanner and scan the INSPIRE centre and surrounding area.
Rob Fitzgerald spoke about the Agriculture Sector Linkages Program in Pakistan (ASLP2) that was co-developed by the University of Canberra and local communities and government in Pakistan. Although ASLP2 didn’t contain any AR experiences, Rob did discuss universal themes such as the importance of engaging and considering learners/intended audience/market/users during the design process and the role of technology in the solution.
Trak Lord (US Marketing and Communications) from Metaio Skyped into ARcamp to present examples of AR in education and also talk about what they are doing in the field of AR and the future of their product. Although it was treated as incidental amongst all the 3D dinosaurs, the 3D object/image recognition feature of the Metaio SDK that enabled the diagnostic/procedural instruction manual augment on the Mitsubishi air-conditioner was my personal highlight. This was the only practical, useful or vocational example of AR as a value-add or assistive tool in his suite of marketing videos.
Amber Standley presented two AR case studies that demonstrated possible implementations of AR. The first AR case study was a marker-based experience that allowed users to learn more about the University of Canberra’s emissions by scanning a poster. The user was then able to:
identify and compare emission levels from buildings that make up the University of Canberra campus
review past emission levels (performance)
display weekly information on buildings with low emissions
display discussions about each building.
The second AR case study was also a marker-based experience that allowed users to learn more about distinguished alumni from the University of Canberra by scanning photographs mounted in an engraved frame. Each photograph triggered a video which displayed biographical and additional information about each distinguished alumni.
Paul Krix provided a very general overview of creating a mobile AR app for iOS or Android with Unity and the Metaio SDK (SDK can be downloaded from the Metaio website). The MetaioMan featured in his presentation.
Danny Munnerley from ARstudio welcomed us all to camp. He spoke about ARstudio as a two-year practical and research project that was nearing completion. He also spoke about the eventual release of a resource that compiles their research findings and that this year would be the last camp.
Tony Oakden and Dean Walshe from the Academy of Interactive Entertainment (AIE) spoke about their experiences developing the 35° 17 South. 35° 17 South was a multi-reality location-based game that took place in April on the grounds of the National Gallery of Australia. Although didn’t contain elements of AR, many aspects of it’s design, development and implementation reflected issues typically encountered when producing technology-based activities.
Design 29: Creating a capital
Alex Young spoke about the issues, development concerns she encountered and insights from developing AR experiences for the Design 29: Creating a capital exhibition at the National Archives of Australia. The Design 29: Creating a capital exhibition brought together the original designs for Canberra by finalists of the 1911 Federal Capital City Design Competition and featured AR elements such as video, animation and detailed scans of original artwork.
The AR mini design challenge was facilitated by Danny Munnerly and Matt Bacon. The challenge was to form a group, brainstorm, design an AR experience on paper and then present the concept to the camp. Prizes were awarded to the best AR experiences. The AR mini design challenge was an awesome exercise in design thinking, where we need to discover, interpret, ideate, experiment and then evolve a concept.
Studio Aurasma and the Aurasma iPad app
Matt Bacon presented a session on designing and developing simple AR experiences with the Aurasma iPad app and the web browser based Studio Aurasma. This session gave us the opportunity to create a simple marker-based video augment with the iPad app as well as explore the features of Studio Aurasma.
Today I attended the second scheduled event for the VET Development Centre Specialist Scholarship. The event was a professional development session facilitated by Greg Stephens. During the session, Greg presented his unique perspective on the following themes that mapped to the nine units that make up BSB51407 Diploma of Project Management:
The People side of Leading Projects
Leading projects in the contemporary workplace
Project Leadership – What’s important?
Leading through the Project life cycle
High performing project teams
Leading teams at each stage of the project cycle
Leading through Project challenges
Addressing resistance; when things go wrong; managing yourself
Greg also initiated lively discussion, activities and presented examples of:
model project success factors such as:
Pinto and Slevin’s 1998 list of 10 project success factors
Turner’s 1999 project drivers diagram
differences between project management and project leadership
project leader qualities
characteristics of effective leaders
characteristics of an effective team
situational leadership model
work preferences – team management systems
Belbin’s Team role summary descriptions
what drives performance and engagement
coaching the GROW model
Tuckman’s four-stage (Forming, Storming, Norming and Performing) group development model with Edison’s (Informing, Conforming and Deforming) expansion
tips for influencing, persuading and resolving conflict
guidelines for leading change
managing employee resistance to change
responding to indifference or anger
a resilience framework (Vision, Determination, Interaction, Relationships, Problem solving, Organisation and Self confidence)
ARIG is a camera rig for recording activity on and around the screen of a tablet or mobile phone screen. The concept for ARIG came from my need to record my experiments with marker and location-based augmented reality experiences.
In this example, ARIG records a simple 3D cube augment produced with Blender 2.62 and Aurasma Studio.
Blender 2.62 does a good job of exporting a 3D scene in the Collada (DAE) format for use as an overlay in Aurasma Studio. You just need to make sure you interpret the newest version of the Aurasma 3D Guidelines in a Blender 2.62 context. For a Blender 2.62 user the most important guidelines to follow are:
Models need triangulated faces (Edit mode > Ctrl+T)
No more than four lamps (lights) although three are recommended
Models are to have no more than one map/texture
Create a .tar archive to upload to Ausrasma Studio made up of .dae (Exported from Blender 2.62), .png (Texture) and a .png thumbnail (256 x 256).
This video is an example of a simple 3D cube augment produced with Blender 2.62 and Aurasma Studio.
As part of my VET Development Centre Specialist Scholarship I’m in the process of developing my practical skills in designing and building augmented reality learning experiences. One of the experiences I’m currently prototyping is a workplace hazard identification activity. This has brought about an interesting challenge. I’m currently grappling with the challenge of using markers placed on the floor to trigger and then engage with an augment containing a 3D object modelled to scale.
A marker needs to be in view and recognisable at all times for the augment to work. An augment containing a 3D object not modelled to scale can be easily triggered and engaged with by a marker placed on the floor as the marker will most likely remain in view and recognisable at all times. An augment containing a 3D object modelled to scale can also be easily triggered by a marker placed on the floor. The user then needs to move away from the marker to engage with the augment. As the user moves away from the marker it no longer remains in view and recognisable. This means the augment will fail.
In this example, a simple augment of an industrial workplace scene is triggered by the marker. The industrial workplace scene has been reduced in size and is no longer suitable.
Increase the size of the marker or place the marker on a wall to ensure the marker remains in view and recognisable at all times. Increasing the marker could be a solution, but then a specialist printer may be required instead of a standard domestic or office printer. Placing the marker on the wall could be a solution, but only if the experience was thematically relevant. A marker placed on wall could also be used to trigger an augment on the floor. This could also work, but would require strict placement to ensure the augment is placed in an accurate position on the floor relative to the marker and not floating in the air or buried in the floor.
Another possible solution could also be to trigger the augment containing the 3D object modelled to scale based on location. This solution could work if the designated location for the augment was outside or if the location could be accurately determined when indoors.