Week 5: Troubles and Presentation

For the past few days we have been working on the green screen/ chroma key feature of ARMedia. But nothing so far has been working! 

Our presentation was hours away and we still can't seem to figure out what the problem was. 

We asked Mr.Mike for help and we were all trying to figure things out at the very last minute. 

It was purely bad time management. We should have been done with this days before the presentation. 

In the end, we found out that ARMedia's chroma key is faulty and is unable to function with it's child which is ARMedia Player. 

Thus we thought if it only works when scanned through the computer's web camera, then it is redundant as  this Horr(AR) Haus needs to use a smart device as it is portable and has the snap camera feature. 

We went on ahead and presented our ideas to the representatives of ROAR Studios. 

Our idea was just to keep the event in one room as a game for the annual Taylor's Halloween night by ADP club. 

The feedback that we received are stated below: 

• Expand the idea throughout the whole campus
• Find something more interesting like a poster that can be scanned to promote the event.
• Since it is a university level, there aren't any high expectations so it's ok.
• Would love to see some prototype for the presentation but it's ok.

Week 4: Discussions/ Locking down the plan

    This week we spent the time in class doing research after research on AR and the types of softwares that are in the market and what are the differences and their pros and con


For today, we've been working on the tracking attribute of metaio. Our issue at the moment is we need something that is fast in terms of loading the animation. To be able to scare people, there cannot be any room for lagging (people will get bored and it wouldn't work as well).
Therefore, cloud-based app are not suitable for this project as it would time to load and play the video. 

Since Metaio is cloud-based, it is dropped from the options. 

The other options that we have are Vuforia, Dfusion and ARMedia. 

We've tried out D'fusion and also ARMedia.
D'fusion has a more complex and advanced feature compared to Metaio and ARMedia that it can be coded for specific functions. It is a little hard to navigate along compared to the other 2 software and I was thinking how it was not relevant to go through so much trouble to figure how it works when another software is able to do the same with little challenge.

Our plan is to model a ghost and then animate to make it seem real. 
Due to this, we have decided to use ARMedia as our preferred software as it works alongside Autodesk Maya which is also our preferred 3D software. 

Where can we get the plugin? 
Inglobe technologies offers the free version of the software from this website: http://www.inglobetechnologies.com/en/products.php

Week 4: Let's Get Started

One week to go until the proposal presentation, we've started to work things out.
Now with the finalized software selection, we are starting to produce the work.

Foong Yann was assigned a task to model the ghost whereas I was given the task to Photoshop little girls to ghosts. This is to be used by Foong Yann as reference.
This is what I have come up with.

Unfortunately I can't find the original picture before I edited but I can assure you the little girl was prettier than this. 

After Foong Yann completed the ghost's head, we went for a progression check with Mr.Mike and he suggested that we do not do modelling instead use a real footage as our modelling skills aren't that great also real footage would make it more realistic. 

A suggestion given is to use the green screen method where we take a green screen clip which are plenty on the web , Youtube etc and use it for the project.

So we proceed on to green screen and I've done some research to test whether is is applicable on ARMedia and it says on the official website that Chroma Keys functions are available and it works just like green screen. Basically, it will remove all the greens in the video clip. 

So that was good news to us. 

Thus, we tested it out. 

Tryout With Environment Tracking

To determine the software to be used, we have narrowed down 2 softwares which are Metaio and ARMedia.

The downside of Metaio that is noticed is that Metaio is cloud-based therefore it takes a longer time to load. Another issue that we would like to test out with Metaio is the the environment tracking and how it works. 

Our plan was to lock down the room and have Junaio (in case we do use Metaio) to load the AR room when the room environment is detected. 

A sudden realization, we have to prepare the room and lock it down in Metaio. Then we have to make sure that the room stays/is arranged exactly as how the room was arranged at first. 

At this, environment tracking is scratched out. 

User Experience Flowchart

To help us to progress, we've come up with the flowchart.

Basically it's the instructions on what the users/ players will have to do.

1. Users will receive a tablet which has an AR player installed in it. 
2. Users will be needed to launch the app.
3. Users will then be demonstrated on how to hold the device and how the AR works. (Users should hold the device at shoulder's length)
4. Users will then enter the AR room. 
5. Once pictures are taken and the adventure ends, users are needed to share the pictures online. 
6. Users will have to wait until the end of the event when we will announce the best picture who will then win a prize. 

That is the rough flow chart for now. Subject to changes later. 

Week 2: Idea development

For this module's project, we were broken down into groups of 3 to work together on one idea for our final project. My group consists of Foong Yann, Ruth and myself. 
Each individual members are to come up with possible AR services to work on with. 

Here are my proposed ideas:

1. Cuti-Cuti Malaysia (AR):

     Basically the idea of this is to make travelling more exciting instead of just going to historical and significant tourist sites like we've been doing for so many years. AR technology can make travelling more exciting. 

    It serves as an tourism app, it gives out information about the sites that the travelers are about to visit before going to the location as well as when they reached and at the same time has an activity that the tourist can choose to try and follow up to. 

   The additional activity involves the tourists to scan any of the historical monument, building or anything of that sort as markers and then launching the AR information regarding it and also a mascot/ character will appear and tourist are able to collect it buy touching onto the character. A collection of these mascots will receive a prize.

  For this idea, the best collaborative partner would be Tourism Malaysia. 


2. Haunted House (AR):

    For this idea I was inspired by the Korean horror comic that I read online. It is a drawing that consists of different frames and readers are to scroll down at a fast motion to be able to see the movement of the characters. 
   
   At one point, the ghost made a jump scare so close to the screen I almost wet my pants. And thus I was inspired. 

  The initial idea of the Haunted House is just jump scares. The players will be placed in an empty dark room and they are given flashlights and a mobile device, preferably a tablet. Through the tablet, they are able to see the world of horror that is in that room. They are then instructed to scan around the room and look for the ghost. Once they are scared shitless by the ghost they are have to then take a picture with it and the best pose will win a prize. These pictures will be uploaded on our Facebook page and people are then allowed to give as many likes. The most likes means the best pose. And the best pose means you stand to win a prize.

  
3. My Pet(AR):

  What I had in mind for this idea is a digital/virtual pet that you can take with you anywhere. Where will it be? In your smart device of course. 

  This idea is not new, but I've wanted to improvise and take the user experience to another level.
With Augmented Reality capabilities, the user can not only see their virtual pet in 3D state but also interact with the pet when it responds to gestures. So you can pretend like you're stroking a cat and it will respond accordingly.