Evaluation

Animation Project Evaluation

For this project, I wanted to develop a way to combine the use of traditional storytelling with new technology. For my physical output, I was able to create two masks based on the characters I had designed and use the masks to ‘bring the characters to life’.

In my project proposal, I pointed out that I wanted to explore the use of augmented reality and use traditional masks to implement the AR. I chose to focus the project on Japanese folklore, using the stories to create various characters which I later conceptualised and created. Within my research, I found very interesting stories on Japanese creatures and found a common trope within a few of them. These creatures were able to shapeshift into humans which I used as inspiration for the characters within my project.

By completing my research and character concept in assignment 1, it gave me an idea of how I want the project to work and look like. I was able to set myself goals and deadlines to complete various aspects of the project in order to finish in time for the final submission. Sticking to my own deadlines have helped the project to progress steadily throughout the year and the crits have definitely given me an insight into the different things I needed to improve and change.

I initially was going to create three different characters which all have the ability to change into humans but cut them down to two. I believed that I had the ability to research, conceptualise and create all three characters in time for the final submission. However, I found myself focusing solely on the fox character compared to the others. Due to this, I was unable to include the cat character within this project. Although I was able to create a physical mask for the cat. I couldn’t complete the preproduction element for the character, resulting in the inability to create the 3D character model. However, with the layout for the show, I found that the layout for two characters seemed to be more visually pleasing and having three characters wouldn’t have fit the space I was given or looked to cluttered. Nonetheless, I would like to research and create this character in the future as I enjoyed the idea of the character. It would also be another opportunity to inform people about another Japanese folklore that they wouldn’t be aware of.

The main aspect of this project is the augmented reality and being able to implement this successfully. At the beginning of the project, I proposed to creating my own AR app but was advised to look at different options because it would be difficult to create an app from scratch. Through my research and experimentation, I found that using Unity with Vuforia was the best option. I knew that this portion of my project would be difficult so I chose to research this within assignment 1. This later benefitted the project because I understood the mechanics of the tool I could easily create the final app for the final assignment.

However, within my experimentation, I used flat images for the AR to recognise to bring the model into space. When I used the mask as the image base the AR would only work head on. This is because the masks are a 3D object and when moving the camera parts of the mask would disappear. I was able to combat this problem by using the forehead of the masks as the image base within Unity rather than the whole masks. The flat surface of the mask allows the camera to detect the surface so when the camera moves in different directions the symbols on the masks will still be visible. Alternatively, I could also paint a symbol on the wall for the final show, which would also act as the image base if the masks do not work. In addition to this, within my research, I found various companies such as Zapper (2011) can generate a QR code that can be printed and used for their AR application.

Although the technique I used has worked, in the future I would like to find a way to upload the model of the masks to act as the image based rather than using an image. This would be able to fix the issue of the camera not recognising the masks because parts of the print have gone off camera. By using the model, the application would be able to recognise all parts of the mask.

As the project is based on combining traditional storytelling techniques with new technology, I believed that finding a way to fix this issue was important because eliminating the masks from the show would fail to convey this message. Therefore, I had to find a solution for this challenge to be able to stick to the original aim of the project.

In addition to this, when presenting my ideas during the proposal pitch I brought in an example of what kind of mask I wanted to create. The mask can be seen in my portfolio, which was made of ceramic. Traditional masks were made of wood or ceramic, I would have liked to continue this through my own project but found that working with plastic is more practical. By having plastic masks, people are able to interact with them without breaking them. However, I possibly would like to create wooden masks in the future as they will look more presentable and professional.

Modelling the characters in Brush was straight forward as I have previously used the software throughout my university career. By looking back at the concept art I had created and various references which can be seen within my portfolio, I was able to create the two characters.

However, texturing the models was a challenge because I have never created my own textures previously. In order to create my own textures, I had to UV unwrap the models which I also had never done. I utilised the UV unwrap tool within the programme but found that it couldn’t handle working with a high poly model. Through research online I found a way to project the high poly model onto a low poly version of the model. I explain this method within the ‘Kitsune Character Modelling’ section of my portfolio. Although this method worked I that there were problems with the UV because the seams would be visible on the face, I was able to use the control painting tool protect the areas of the models from forming seams.

I also experimented with texturing in substance painter but found that the texturing was pixelated because of the low poly model. I tackled this by texturing the model with polypaint. Using this method was the better option as I was able to view all the sub tools together and have a uniform colour pallet as all the parts were visible. Again, I explain this in detail within my portfolio.

In addition to this, I found animating the characters challenging. During the crit sessions, I was suggested to animate the characters with idle animations such as breathing. I purposely chose not to animate at the beginning of the year as I felt that this was not my strong suit. Nevertheless, I decided to try to animate the fox and raccoon characters I had created. I successfully rigged the characters with the auto rig tool in Maya but found that it deformed the mesh when I tried to animate the character I found that the deformation of the fox character could easily be hidden with clothing as it was mainly evident on the body. However, as the proportions of the raccoon was not human-like, it causes the whole mesh to break which I highlighted within my portfolio. Due to this, I have decided to not animate the characters because I was not able to figure out a way to animate the characters without breaking the mesh. In the future, I would like to rig the characters from scratch and see if that will fix the deformation of the characters mesh. I do agree with the feedback I have gotten, suggesting that the animations would make the characters more lifelike and improve the overall project itself. As the project is a prototype, I believe that having the characters in an A-Pose or T-Pose will still look professional.

Finally, I was able to successfully create the clothing for the character within marvelous designer (MD) by using patterns I found on the MD forums which I have referenced below. I have never used the programme before but found that it was quite straight forward to use. I initially experimented with MD without a pattern and found it difficult to make the clothing as I don’t have knowledge on how clothes are made. Using the pattern gave me a basic kimono which I later adjusted to fit the proportions of the characters. Overall, I didn’t find any problems with the modelling of the clothing but I would have liked to make more complicated clothing for the characters in the future. I decided to give the characters basic clothing so it doesn’t take away from the characters itself, but I would have like to create accurate kimonos for the characters to accurately depict how they would be dressed in the stories I have researched.

Although I consider my project to have been successful, there are elements I would perhaps change, should I be given the chance. I would definitely include the third character I originally wanted to create into the project. The character fits the theme of shape-shifting anthropomorphic characters and the project would have benefited from a collection of more characters. I also would have liked to create the masks out of wood or ceramic which is what masked were traditionally made of, similar to the mask I had previously presented within the project proposal pitch. This could make the project seem more professional because of the delicacy of the material of the masks. Finally, I would have liked to animate the characters to not only have an idle pose but to have more complicated animations. This will depict the character’s personality which may not translate through the final year show. However, I want to treat this project as a prototype of how the application would work and look like. This project has definitely been able to cooperate with a traditional storytelling tool of the masks with advanced technology. Hopefully in the future, I will be able to develop this project further to create an augmented reality application that people can download and use.