Semester B: Week 29
Student: Tai Ser Yeet (22064351)
Programme: BA (Hons) Digital Media Design
Task: Week 29's Class Task
WEEK 29
This week, I went to Steve for a tutorial session to discuss my project prospectus.
Feedback from Steve
- Better to use Microsoft Word compared to Google Docs because you can make the pages horizontal and vertical
- Think realistically about how you can produce this independent project
- This AR app with 3D modelling sounds like a degree project because it is a lot of work especially since I have never used Blender before so restructure your idea again, think if it is feasible within the stipulated time frame
- You can try Pro Builder in Unity if that is easier
- For the independent project, you do not have to fully develop the idea because Steve just wants to see the skills that you obtained from the module. I could even complete the project over the summer.
- During the summer, try to pick up useful skills such as Blender which will be very useful in the future.
- Compile a list of companies you want to call up for internship opportunities and find out the HR’s name and number. Call each of them one by one.
- Steve said I do not have the necessary skills to carry out such a project although he still thinks that this project is really good
- Steve accepted the example of the mood board I gave and thinks it is good so I should make something like that
- The cultural context section is good but I should explain more about why I am interested in this topic and perhaps say how it could benefit me (it can be unrelated to my project)
- Include more direct quotations from articles talking about AR and education, remember to insert citations as well
- Target Audience: It is a definite no-no when your target audience consists of an 8-year-old and a 28-year-old because their behaviors, wants and needs are totally different. If you think about it, you don’t see Instagram catering to adults and children.
- You can just have primary school kids as your target audience and why you chose them
- The equipment section is okay
- Production Timeline: The lines need to be color-coded because all of them are under different categories
- The processes should be more detailed e.g design and wireframing should be under research and idea development
- Risk Assessment should be in the template that Steve gave
Feedback from Steve
- Better to use Microsoft Word compared to Google Docs because you can make the pages horizontal and vertical
- Think realistically about how you can produce this independent project
- This AR app with 3D modelling sounds like a degree project because it is a lot of work especially since I have never used Blender before so restructure your idea again, think if it is feasible within the stipulated time frame
- You can try Pro Builder in Unity if that is easier
- For the independent project, you do not have to fully develop the idea because Steve just wants to see the skills that you obtained from the module. I could even complete the project over the summer.
- During the summer, try to pick up useful skills such as Blender which will be very useful in the future.
- Compile a list of companies you want to call up for internship opportunities and find out the HR’s name and number. Call each of them one by one.
- Steve said I do not have the necessary skills to carry out such a project although he still thinks that this project is really good
- Steve accepted the example of the mood board I gave and thinks it is good so I should make something like that
- The cultural context section is good but I should explain more about why I am interested in this topic and perhaps say how it could benefit me (it can be unrelated to my project)
- Include more direct quotations from articles talking about AR and education, remember to insert citations as well
- Target Audience: It is a definite no-no when your target audience consists of an 8-year-old and a 28-year-old because their behaviors, wants and needs are totally different. If you think about it, you don’t see Instagram catering to adults and children.
- You can just have primary school kids as your target audience and why you chose them
- The equipment section is okay
- Production Timeline: The lines need to be color-coded because all of them are under different categories
- The processes should be more detailed e.g design and wireframing should be under research and idea development
- Risk Assessment should be in the template that Steve gave
While I was there, a tutor from Level 4 also gave a few comments on my app ideas.
Feedback From Level 4 Lecturer- Begin searching for object detection
- Try this software called ml5.js
- Watch Coding Train to learn more about javascript
While I was there, a tutor from Level 4 also gave a few comments on my app ideas.
Feedback From Level 4 Lecturer
- Begin searching for object detection
- Try this software called ml5.js
- Watch Coding Train to learn more about javascript
Example Given By Peter (Level 4 Tutor)
Peter showed an example of object detection using Augmented Reality which shows that the software can identify whether the object is a human being, an animal, or a regular piece of thing. However, this software is not easy to code, and he expressed that even with his level of expertise, he needed at least two weeks of continuous work to program this feature. That scared me a little and gave me a harsh reality check on how much time I had to complete this project. Not only that, Steve told me upfront that I do not have the necessary skills at this moment to execute such a heavy project. Steve also advised that even though this is an excellent idea for a project, he thinks it is more suited for a degree project instead. Fortunately, Peter gave a few suggestions on other methods I could explore for object detection including the integration of markers or the use of images to make the 3D model appear. Research also states that the "marker-based approach is the type of AR application that was most frequentlyemployed in both primary and secondary education." (Basumatary and Maity, 2023)
To determine the best solution for my project, I also analyzed the cons of using marker-based targets. Technical issues might arise due to the difficulties in recognition of image targets, occlusion, the requirement of the AR applications, and consuming high power and memory. Usability issues encompass the requirement for a better user interface (UI) and difficulties in handling the markers to produce the required 3D target images.
Example Given By Peter (Level 4 Tutor)
Peter showed an example of object detection using Augmented Reality which shows that the software can identify whether the object is a human being, an animal, or a regular piece of thing. However, this software is not easy to code, and he expressed that even with his level of expertise, he needed at least two weeks of continuous work to program this feature. That scared me a little and gave me a harsh reality check on how much time I had to complete this project. Not only that, Steve told me upfront that I do not have the necessary skills at this moment to execute such a heavy project. Steve also advised that even though this is an excellent idea for a project, he thinks it is more suited for a degree project instead. Fortunately, Peter gave a few suggestions on other methods I could explore for object detection including the integration of markers or the use of images to make the 3D model appear. Research also states that the "marker-based approach is the type of AR application that was most frequently
employed in both primary and secondary education." (Basumatary and Maity, 2023)
Research on Markerless VS. Marker-Based AR With Examples
Research on Marker-based AR AppThe marker-based AR requires the user to scan a marker that triggers the augmented experience (whether an object, text, video, or animation) to appear on the device.
What are the criteria for an element to be considered a marker? - Images with lots of corners and edges work
- The object has to be distinctive and recognizable for the camera to properly identify.
The position of digital elements also depends on the location of the marker. In most cases, the experience will display on top of the marker and move along with the marker.
Research on Marker-less AR AppThe marker-less AR launches the experience by selecting the AR feature in an app or website or clicking a link. The user scans the real environment and places digital elements on a recognizable feature, like a flat surface. The models are in a way, not tied to a certain location.
Creating the Low-Fidelity Wireframes
To visualise the app better, I dived straight into designing the rough prototype for the app in Figma. I also decided on both marker-based and marker-less AR technology. For the marker-based technology, users would need to scan an illustration of the oyster printed on a square piece of card stock. Upon scanning, a 3D model of the oyster is complete with floating labels on each of its body parts. If the users would like to explore further into a specific part, they could do so by simply pressing on the label and an information card will pop up.
To educate the users about the oyster's lifecycle, the app requires them to place the 6 cards representing each stage the oyster goes through in the correct order. After completing, the system would show the accurate answer on the screen.
Users also have the option to reposition the 3D model into the scanned surface. The ability to interact with the model including rotating and scaling is provided.
I also included safety warnings, privacy policy, and instructions in the design.
Click-Through of Low-Fidelity Wireframes
Research on Markerless VS. Marker-Based AR With Examples
Research on Marker-based AR App
The marker-based AR requires the user to scan a marker that triggers the augmented experience (whether an object, text, video, or animation) to appear on the device.
What are the criteria for an element to be considered a marker?
- Images with lots of corners and edges work
- The object has to be distinctive and recognizable for the camera to properly identify.
The position of digital elements also depends on the location of the marker. In most cases, the experience will display on top of the marker and move along with the marker.
Research on Marker-less AR App
The marker-less AR launches the experience by selecting the AR feature in an app or website or clicking a link. The user scans the real environment and places digital elements on a recognizable feature, like a flat surface. The models are in a way, not tied to a certain location.
Creating the Low-Fidelity Wireframes
To visualise the app better, I dived straight into designing the rough prototype for the app in Figma. I also decided on both marker-based and marker-less AR technology. For the marker-based technology, users would need to scan an illustration of the oyster printed on a square piece of card stock. Upon scanning, a 3D model of the oyster is complete with floating labels on each of its body parts. If the users would like to explore further into a specific part, they could do so by simply pressing on the label and an information card will pop up.
To educate the users about the oyster's lifecycle, the app requires them to place the 6 cards representing each stage the oyster goes through in the correct order. After completing, the system would show the accurate answer on the screen.
Users also have the option to reposition the 3D model into the scanned surface. The ability to interact with the model including rotating and scaling is provided.
I also included safety warnings, privacy policy, and instructions in the design.
Click-Through of Low-Fidelity Wireframes
