Individual Reflections - Catalin

 1st assignmen

    For this assignment we did a geography based application in which the user displays a flag to the camera and on the screen, a map with the respective country highlighted on it. This assignment was Markerbased-AR focused, meaning that we had to combine reality and specific virtual components. In this case, the source of the flag (a piece of paper, an image on the phone) is the real-life object and the map-flag relation is the virtual one. 

    The project's main technologies are image tracking and recognition. This is done by mapping each flag to a copy of mentioned map with each country highlighted separately. These mappings are then put in a list and used as described above.

   Regarding issues we encountered during this assignment, the worst would be that the image recognition did not work properly on simple flags like France or Germany, it sometimes worked with bit more unique like the Nordic-cross flags and it always worked on distinct flags like Albania or Andorra. During development we fluctuated from thinking we implemented the feature wrong, to just accepting that it might have tracking issues due to light or other external factors. This assignment being small in scale, all three of us, me, Alexandru and Roksana worked on the same things at the same time, from research to drawing maps and putting them in unity, to testing. 

2nd assignment

    For the second assignment, we wanted to continue with the educational themes and we decided to create a small Markeless-AR museum about the evolution of soldier attire through history. For this we needed a lot of diverse assets and that turned out to be the first problem. For the amount of characters we wanted to present, we were unable to find enough good free assets, so we then filtered them a bit and we were left with 3. For this assignment I mainly researched some useful information to display in this museum, and searched for assets. 

    Sadly, we had a lot of problems with this assignment, as we were having a hard time rendering the location and the character assets properly. We tried any solution we found online but the result was always the same. Because of that, we tried to first spawn a portal to this location and we hoped that will fix the issue, but then other types of problems started appearing, the worst being shader issues. After switching to the portal idea we spent most of the time trying to fix that but in the end we presented it using gifs of how it should play out, using Unity's in-game window to show that the camera is seeing. 

3rd assignment

    I believe the third assignment turned out the best from all points of view. We liked the idea and we agreed on all points on how to proceed. This assignment was a VR game about being a waiter in a restaurant. People would spawn in and the player's job is to deliver the food that is specified in a text field above their head, disappearing afterwards. Food would behave almost the same way, spawning a random food item after the previous one has been picked up, creating a sense of urgency in case you don't have the correct item on you. 

    There weren't that many issues with this one, besides the occasional bug. I'd say we had more issues with the version control, sometimes resetting our progress when pushing and pulling. Until we got the hang of it, this was the biggest nuisance who in the end costed us some time. This means that we were not able to create a proper end game like we initially planned. 

   Another last issue that affected this assignment was the connectivity and reliability of the VR headsets who most of the time did not connect, or only worked only on one laptop. We later fixed this issue, but this being the first VR application, sadly it suffered the most. 

4th assignment 

    This last project was a mixed bag. The idea at the start barely resembles what the final project ended up being. This doesn't mean that we as a team are not satisfied with the final version, but looking back at what we wanted to do, I'd say we were way too ambitious for our experience. 

    For this assignment we chose to go with something similar to the third assignment, but we were having a hard time figuring out something better than that one that was doable in the given time frame. From the original maze filled with puzzles that are solved with gestures, we ended up with a fast-paced-gesture-recognition-to-get-a-gun minigame, which at least worked properly.

    The biggest challenge for this project was locomotion, which we finally fixed in the last week. We were happy at least that the game can be played fully without any controllers and movement and object interaction are working, even if we got rid of the puzzles.