1. Action Points
1.1. Develop the product and marketing strategy with Tango
1.1.1. Integration with a Deakin AR app adaptation for Theme parks to be shown in November at IAAPA https://www.google.com.sg/?ion=1&espv=2#q=amusement+park+event+orlando
1.2. Develop the support of Tango's capability within EON engine
1.2.1. Node and engine feature
1.2.1.1. Capability
1.2.1.1.1. AreaLearning
1.2.2. Prototype and app feature
1.2.2.1. Occlusion support with Depth map and Shader
1.3. Develop stun application for demostration and proposal
1.3.1. Themes
1.3.1.1. Smart Nation
1.3.1.1.1. Smart Navigation (fire fighting)
1.3.1.2. AVR classroom
1.4. Comparison between hololens & tango
1.4.1. Jan: Set up a comparison chart between HoloLens, Intel RealSense and Tango (and more if applicable) to understand the technical limitations of Tango compared to more high-end systems like HoloLens.
1.4.1.1. Jan: Create a brainstorming document with ideas of how we can best use Tango when we better understand the limitations.
2. Ideas
2.1. Develop a AR UX prototyping Tool such as POP or InVision_ See proposal here: https://drive.google.com/drive/u/0/folders/0B4-3TjSkDRk7Ml80cUw2enBOc0U
2.2. Idea 2
2.3. Idea 3
3. Prioritized Ideas
3.1. High Priority
3.2. Medium Priority
3.3. Low Priority
4. Goals
4.1. Develop the proposal with Google on Smart Nation's theme
4.2. Develop the support for AVR product
4.3. Allow other EON developers who is interested with Tango to develop their idea with EON SW
4.4. integrate Tango For White Label Deakin AR App
5. Problems
5.1. It's difficult to associate the information with the physical object without special settings to the physical object (place marker on top of object)
5.2. We need marker for display virtual object and always forgot to grab the marker
5.3. Problem 3
5.4. Testing of Limitations
5.4.1. The tracking is bad/faulty when looking at reflective surfaces. Reflection affects the depth sensors
5.4.1.1. How to test : Look at the floor while walking in an VR environment. Take note of distance moved. Next look at a reflective surface and walk forward. You will notice u moved less while looking at a reflective surface.
5.4.2. Pointing to sky/high ceiling will cause the app to lose tracking and when righted. The app will attempt to resync but will continue to drift
5.4.2.1. best to have a reset location feature button or automatically called after such an incident.
5.4.3. Sun infrared will totally not work at all. The constructor will build in sheltered areas but when going to the unsheltered area the scene is completely blank.
5.4.4. Too little contrast Having the depth sensor and video feed focused on a pure colour will cause the app to think it is going forward indefinitely.
5.4.5. Distance too far Moving away for 4 meters will create a 0.2m error based of measuring app. However it was in semi outdoor area and the error margin did not look like 20 cm.
5.4.6. Shaking the Device too much makes the tracking go crazy and the app will hang while it processes its bearings.
5.4.7. Moving too much/fast will create a lost of tracking. This will cause the points to be off by a error margin.