3D Class S3 - Personal Project 02 Final

So here is my final corner piece that I was able to complete in 3 weeks, I will continue this in VR and will have a more fully completed environment! This was a really fun environment for me and hope to improve on it!

Software Research (Marvelous Designer & Agisoft)

During my time in the third semester at FIEA I wanted to dive into 2 programs that were new to me, the first one being Marvelous Designer which is a program that is used to create 3D virtual clothing or fabrics that can be draped over an object or character. My reason for using this program was to create curtains, table runners and cushions for my hotel hallway scene.

At first you have to become familiar with navigating around the space which is somewhat similar to Maya. The best way I was able to learn the program was to follow some tutorials on YouTube because there are so many ways to create what I wanted for my scene. The first thing that I wanted to do was to create curtains but instead of just making them in the scene I wanted them to naturally form with the environment placed in so I exported out the pole, furniture and window sill combined as an FBX and imported it in Marvelous Designer.

Once in the scene I created the plane for the curtain and checked the scale of it with the environment so it could match the scene scale wise. I also simulated wind so it could push the curtain and make certain folds and make a better silhouette than just boring curtains hanging down. If I wanted to I could also pull the curtain itself and change it that way but then it could also mess up a little. One way to attach the curtain to the pole was to create a point and then pin it to the desired mesh, I wasn’t too fond with the method so I ended up looking up for another video that showed me how to freeze the curtain in midair without it falling to the ground when simulating the curtain. What I did was to create a small strip of cloth and freeze it in the space, once doing so I created another longer piece which was the curtain itself and sewed the two together. The result was very nice and you have control of the length of the strip so you can create nice folds for the tips that looked very natural. You also had the ability to pull either side of the strip so you could change the folds at your desire.

The next thing I wanted to do was to create a pillow for my chair and that was also a fairly easy, first I had to create two squares and align them correctly, I then had to select the matching edges before sewing them together so they would form correctly. Once the two pieces were together, you select on plane and on the right side you will see a property editor and go to the pressure option and first make it a negative value and then select the second plane and do the same thing on the pressure option but this time make it a positive value. Once you have that done you simulate the pillow and you will see how the pillow inflates. From here I used the grab tool and placed the pillow on my chair so it could fold itself depending on where I placed it. If I wanted to I could’ve made ruffles for the cushion but because of scope I decided not to and go with tassels.

The little that I learned from using Marvelous Designer, I realized that how important this tool helps in the industry and how much it’s been used to create clothes for characters such as Metal Gear Solid, Assassins Creed, black Desert, Maple Story and the Witcher just to name a few.  

The second software I wanted to try out was Agisoft Photoscan which is a photogrammetry program that uses digital images and generates 3D models. After realizing that this method is used a lot in the industry I wanted to try it out and see what the end result was so I partnered up with Chelsea Zui and we tested out several things. The first thing we did was to photograph a funko-pop but we took about 12 images. After putting in the images in Agisoft, the program only recognized one image because of multiple factors. One of them being the lighting was too bright in the background, we needed way more photos from different angles as well in order for the program to collect more information in order to build out the cloud particle before generating the 3D mesh.

The second attempt we ended up photographing a bust sculpt and we ended up with around 80-90 images from various angles. At first we attempted placing in the images inside of the program and it kinda work but what Chelsea and I ended up looking up was that we had to take the images in photoshop and actually mask out the object from the background. Once we did that we took it back to the program and re-did the particle cloud again and the result as a lot better than the first. We also had to clean up the particle cloud in order for the mesh to generate properly without any floating points. Once the mesh has been generated at the quality you desire then we can import that mesh into ZBrush to do some clean up and then exporting it out to retopologize it in Maya.

                I didn’t go much further from what I just wrote on here about the software but from the little that I learned from the process, it showed me how we can generate a 3D mesh from just images. Even though we don’t have a 3D scanning gun like EA, we can still use other programs that can help us get what we want even though we have to go through a few more steps in order to achieve the end result.