In my Imagined Worlds group project, I am required to make the model of the robot which is the main asset in our animation. I am also required to texture it, rig it and animate it. I have created some concept art for this model before setting on making the model. I have managed to make the low poly and high poly models; however, that is as far as I have gotten due to having some issues using the curves tools as well as naming around 66 body parts, parenting parts together and group along with setting a UV map layout for a set group of body parts.
In order to complete my task, first, I must import the low poly model into substance painter so that I can start to back the high poly model onto the low poly. Then I can start texturing the model. Once that is complete, I can apply the texture file to the model in Maya and start to work on rigging the model so that I can start to work on animating it.
In order to complete my work, I need the other assets models from my team mates so that I can get my model to animate around the environment as well as interact with some of the models. I need objects such as the table so I can animate it laying then getting up when the boy is finished making the robot.
Looking at my low poly (left) and high poly (right) models, I think that it has a toy like robot look right. The issue I had with tris has been fixed so there are only quad polygons in the model. I may need to figure out how I am going to animate the eyes since I think that it will be texture animated instead of having a model animate. That will be something that I will have to look into. I think that the main issues that I had making this model has been fixed such as naming the body parts and parenting everything.
I could look into some tutorials to see how I could get textures to animate as well as any substance painter tutorials that may help improve the look of my model.
I am working on a group project called Mechanical Bonds. This is part of our Imagine Worlds project. Me and my group are working on a 3D animation showing the relationship of a boy and a robot over a period of a 30 second long animation. I am currently working on modelling, baking, texturing, rigging and animating the robot. At this current point I have made the high and low poly version of the model that I will use for the animation.
This is the low poly version of the robot model. I have applied different materials to different parts of the body. The UV’s of the sections in one particular colour have been grouped and have the UV map layout together. So the right arm has its own UV map layout to make texturing easier along with the leg and main body having their own UV map layout. I have also parented the body parts together with the main body being the main parent and named each individual section. Around 66 polygons that I had to name individually.
These are my high poly models of the robot. I have added more detail by extruding some of the faces in and bevelling some of the edges. The quality of my model and research has been reviewed by another person to get their thoughts on my model, areas of improvement as well as my research into making the model.
Peer Review. This is the review that was made.
Within my current lesson, I have been trying to get to grips with Substance Painter. Substance Painter is used to texture 3D models more easily than other software’s such as Photoshop since both the UV map and the model is shown so you can see how the model looks whilst working on it.
This is the image of how my go at the turtle model currently looks (the model was provided, the texturing I did to get to grips with it). I started off colouring the shell with the polygon fill tool. This tool can be used to fill the entire part of a model (which is useful when wanting a part to just be on colour like a white shirt) or have use UV fill to colour as you may see by the turtle’s shell. you need to create a fill layer for the fill tool to work however.
The fill layer also allows you to add a black mask to the fill tool. A mask makes the colour not appear. There is a slider that affects the visibility of the colour, white means that the colour shows on the model and black resorts the model to its original colour.
The UV fill allows you to fill the UV shell in so to it gives a part of the object the colour that it needs to be filled on the map to the right. Polygon fill allows you the select and fill in polygon faces of the model.
You can also change the amount of light reflection on the fill so that they will have a metallic or plastic look with high light roll off to bounce off and make it shiny or make it look like more cloth with less light reflecting off the object. You can also apply materials on to the object, so you can apply textures to make the object such as the turtle shell have the wood texture to make the shell wood. You can download materials off the internet and import them in to use. You can also apply logos onto the model. There are some logos already available but you can download and install more. The best part is that you can apply the logos to the model itself instead of the UV (which you can still do) which makes applying logos more easier to be precise with.
Substance Painter is a better option for texturing compared to other programs such as Adobe Photoshop simply due to the fact that you can see the model in real time whilst texturing so you can see if the logos are in the right place, if the colouring is correct, how it looks and change it where needed rather easily.
Within college, we have to work on a group project where we select which out of the six options that we want to work on. I chose to work on the second option which was create a 3D animation showing the relationship between two characters. I was put in a group of 4 people including me who chose to do that option as well. We are working on the relationship between a boy and a robot that he built which develops over time. We named the animation project Mechanical Bond and organise the things that need to be worked on, which roles that we take, an other content that will help us such as contact numbers, mood boards and other images on a website called Trello.
Alex H works on the storyboard of the animation and researches on how we could show the relationship between the boy and the robot and is the leader. Alex T works on the character model and rigs it whilst working on his animations. I work on the concept art for the robot along with modelling it and animating it. Wesley works on the some of the environment objects along side me and Alex H. We have until Christmas to work on this animation, work to make sure that everything checks out.
I am currently working on a personal project of making a 3D model on an animatronic in Maya. The animatronic that I am currently working on Circus Baby from Sister Location of the Five Night’s at Freddy’s series. I had made other projects after I first started college working on other animatronics from the series that was passable for the first attempt; however, they ended up looking flat, not very detailed and I didn’t know how some stuff worked so holes for the eyes ended up being a black cylinder at first.
This is currently how far I am into the modelling process. The eyes being coloured in doesn’t mean anything, it was just easier to colour them in before I started to combine everything plus it adds a bit of life to the model whilst working. I had imported an image of my drawing for reference so that I knew where to put the wires. If you are going to import images into Maya, select the view by pressing space then select and import it into the front, side or top view, not perspective as scaling will be problematic if you import in perspective mode. To import, just press on view under where you choose the polygons to make the model, select image plane then import image.
The next part is just trying to make the model look as close as the reference as you can make it. The eyes is where I started since it was the more complex part of the model. I used a sphere for the eye ball an used a cube for the eyelids. The cube was broken up into segments so that there was more edges that I could move the edges to make eyelids, the eyelashes and wires are both made out of cylinders. Extruding is required to make the model more realistic.
This is my current progress on the model. I still need to make the face plates more round as well as sort out the hair so that it fits the top face plate along with including the outer parts of the hair. The edges also need to made smooth to improve the model. I should also texture the model. Getting the hair to work on the face plate has proved challenging as I need to Booleans difference so I get the gaps but the models aren’t properly lined up. I will need to rotate the model around since rotating the camera isn’t really useful in this situation.
Over the corresponding weeks, I have been practicing making low poly models for a Wacky Races style race track that my team was tasked with. We had to create a small part of a race track with at least one vehicle featured in it. Each model had to be under 1000 tris (triangle faces). My job was to work on the vehicles. I also worked on the trees for the environment. The environment wasn’t my main job but it was something that I assisted in. I started in making concept art for my vehicle designs before actually getting to work on the models.
The sports car was inspired by Sonic’s car in the Sonic & All Stars Racing games. I didn’t want it to look basic so I included some patterns to it with headlights acting as eyes. The model consists 474 verts, 1056 edges, 576 faces, 960 tris and 1020 UVs. I used a fair amount of booleans, extrusions and triangulate/quadrangulate in making may model.
This vehicle was based on the monopoly car. I included the top hat as well a funny little moustache and eye glass on to the car to give it its own personality. I have also included some red and white lines as well as the words ‘Monobile’ on the sides (Unfortunately, I wrote Monobile backwards). This vehicle wasn’t as big as the other car with 398 verts, 855 edges, 470 faces, 778 tris and 918 UV’s.
The final piece of modelling that I worked on was the tree. I made this out of three shapes: two cylinders and a sphere with the amount of faces decreased. The design is rather basic, I mainly just moved the vertex’s out to make it an imperfect shape. I used a ramp shader on the leaves to make a cartoony look to the tree. The model contains 191 verts, 410 edges, 221 faces, 378 tris and 360 UV’s.
I will be covering the different types of mapping available when modelling. Bump maps can create detail in the model that is actually fake. It doesn’t add any additional resolution to the model as bump maps are grayscale images that are limited to 8-bits of colour information. The values in a bump map are used to tell the 3D modelling software up or down. If the values are close to 50% grey then there’s little to no detail that will show through on the surface. Turning the colour closer to white will make the details appear to be pulling out of the surface whereas turning it closer to black will do the opposite and make then appear to be pushing in. Wrinkles and pores are some examples of what bump mapping can be used to create. Bump mapping can be rather easy to create and edit in a 2D application such as Photoshop or Illustrator, however, bump maps can break if the camera views it at the wrong angle due to the detail being fake. The silhouette of the geometry that the bump map is applied to will remain unaffected the map.
Normal maps can be referred as a better type of bump since it has essentially replaced it. like bump maps, the detail that they create is also fake. However, unlike a bump map, normal maps use RGB information that corresponds directly with the X, Y and Z axis as the information tells the application the exact direction that the surface normals are oriented in for every polygon. The orientation also tells the application how the polygon should be shaded. Normal maps can be rather difficult to create or edit within a 2D application such as Photoshop so a likely option would be to bake a normal map using a higher resolution model.
Displacement maps are king when it comes to creating additional detail for low-resolution meshes. In order for detail to be created based on a displacement map, the mesh must be subdivided or tessellated so real geometry is created. Displacement maps can be painted by hand or baked from a high resolution model and it consists of greyscale values like bump maps. 8-bit displacements are available although they aren’t the best option as using 16 or 18-bit will provide a better result since 8-bit displacements may look good in a 2D space but can sometimes cause banding or other artifacts when they are in 3D due to insufficient range in value.
Source of information: https://www.pluralsight.com/blog/film-games/bump-normal-and-displacement-maps.