Here is a look at the proxy head and the bank notes. There are 82 bank notes in total, each was separately placed one by one. It did not take me as long as it sounds it would have haha.

I was given a diagram showing the lighting setup in the original photograph which I replicated inside Maya using a few spot lights and final gather. Below is a close up shot of the final render.

I Managed to model the cloth around his head and his arm straps relatively fast because they are quite simple.  The wings took a little longer because there is a lot of detail where the wing meets the body.  I like to poly model as much detail as possible before I make displacement and normal maps.

Here is a Hi res turntable render of part of the skeleton and an example of one of the colour maps used:

here you can see that I started texturing the muscles. This is the first pass for everything texture wise. Once I have blocked in all the textures I will then go in and detail everything further. Such as add blood stains on the bones around the muscles, fungus on the bone and muscle, More areas of muscle that are rotten and gross looking. it’s a lot of work but I look forward to it.

I looked at a huge amount of reference from anatomy books and websites. I also got a lot of help from VisibleAnatomy.com This was a great way to properly learn about anatomy. I have gained a huge amount of knowledge as a result and I feel even more confident In coming up with ever more complicated character designs. now I can spend more time on refining the look and style of my characters and less time trying to figure out if something is perfectly anatomically correct.

Modelling the skin has been very tricky. I would begin a body part by putting in basic edgeloops and then start to extrude the edges and adjust the verts. The model can become very detailed and dense very quickly but I am finding it a lot easier to handle. I model the skin with the muscles and bones visible so I can be more accurate with the skin. The character is supposed to be a type of zombie in essence so the body is partially decomposed, the skin is tight etc. I don’t want the skin to be too tight so all the muscles are very defined because it is not the look I am going for, and I think the final render will be semi-transparent so the muscles underneath will be visible.

Here is a look at the feet. I wanted the skin to be all cracked and ugly looking. The character does not wear any proper shoes so it would make sense to have really worn out feet. I wanted them to look like he had been walking for a long time in a wet/swampy kind of environment so his feet would be swollen and wrinkled.

I liked the idea of modelling torn skin and muscle. There is a chance to really see how much detail I could put in with only poly modelling. Of course I will need to make it asymmetrical when the model is done. Otherwise it will just look silly to have perfectly symmertrical bits of ripped skin. I also need to detail the face more, especially around the eyes and nose. I think I may also model part of the gullet to make it extra disgusting. I would say the model is about 75% finished at this point. the final 25% is detailing and tweaking, not including making the wings and arm straps.

After I decided on a name I needed to design the bottle.  I used photoshop to create silhouettes of the bottle.  In all areas of design, whether its a fragrance bottle or a character for a film, the silhouette is the most important aspect of the design because it can quickly convey the tone of the design.

I wanted a very strong, bulky looking bottle to match the strength of the alcohol. I ended up choosing Bottle number 6. It was not the most creatively designed bottle out of all the silhouettes but I think it fit what I was trying to accomplish. It was quite simple to model. I found that complicated and artist bottle designs are only really seen for fragrances.

Once the modelling the bottle was complete I moved on to design the the label. This was done in a similar fashion to the designing the bottle. I made a silhouette design in photoshop and the modelled it in maya. I modelled the label by duplicating some of the polygons on the front of the bottle and I made a cookie cutter type shape from the silhouette design which I then used as a boolean to cut the shape out of the duplicated faces. This insured that it kept its curved shape around the bottle. The label shape was then smoothed and Uv’d.

The next step was to render it. I had a simple setup. I used dielectric materials for the bottle glass and the liquid. The lighting was rather simple, A couple of key lights, a HDRI dome and a few planes with white surface shaders on applied to them for studio reflections. I then put the render into photoshop and created the background and smoke effects.

I am quite happy with how it turned out. It took me around 2 and a half days to complete. I did not have a clear idea of what it was going to look like from the beginning and it would have been done quicker if the lighting scene and design of the bottle and label were already worked out.

I finished the hairless rodent project!  I have been working on multiple projects at the same time over the past few months and I am glad to see that this project is completed.  I am quite pleased with how this has turned out and I am happy to have learned as much as I did about the whole process.  The compositing phase of this project was hard to get to grips with.  I know a lot about the 3D aspects of this kind of thing but I found it hard to adjust while I was compositing in 2D.  Suddenly I was dealing with shuffling image channels, rotoscoping and marker removal.  This was hard to get to grips with at first but after a while I started to think like a compositer and I could overcome problems that came up.  Do take a look at the finished shot and the breakdown of it in the video below.

Case Study

 Rigging

After I had modelled and textured the rodent the next stage was to rig it.  The most difficult part of rigging the rodent was skinning.  It was very hard to try and give it the look that the skin was loose and stretchy.  I did not take advantage of Maya’s new skin and muscle system like I first planned, purely because I wanted to see if I could get good results by using normal skinning techniques and I wanted to use pose space deformations. you can see in the picture below I also rigged the skeleton. This turned out to help me quite a lot when skinning because I could imagine where the muscles were attaching and if the bones would pertrude through the skin etc.

I used a few scripts that I found on highend3d to speed up a few of the really tedious tasks. One of which was the “creat fk controller” script. This simply created an fk control on a joint with all the transforms zeroed out. Doing this manually would have been very tedious and boring as there are many controls on the hands.  One set of controls I always like creating are the eye controls.  Specifically the “fleshy eye” setup.  Using blendshapes and set driven keys I could create the look that the eyes were tugging on the skin of the eyelids as they rotated.

Pose Space Deformations

A skinning technique that I was really excited to incorporate in this rig was the use of the amazing poseDefomer plugin by Michael Comet.  To briefly explain what pose space deformations are, they are corrective shapes that are activated based upon the pose of a joint or group of joints.  For example, on the rodent’s rig the skin had to be able to wrinkle in certain poses and be smooth in others.

As is shown in the videos to the left, parts of the skin deforms and wrinkles as the area in question is compressed.  These little deformations can be quite subtle and if done right they arent noticed at all but would be missed if they weren’t there.  I made many facial and body shapes that were for correcting areas that lost volume like the area below the jaw, or for areas that needed to wrinkle such as the neck and torso.

Animation

I dont have as much experience in animation as I do in other areas of 3D but I am still quite pleased with the animation, because I have not had that much practice at character animation and I think it is good for someone so inexperienced.

HDRI Lighting

I got to try out my mirror ball and create a HDR panorama so I could create a nice indirect lighting pass for the rodent.  It was easier and less problematic to create it than I thought it would be.  The workflow to create it was fairly simple.  Place the mirror ball in the area where the rodent was and take multiple pictures of it at different exposures.  Then I imported those pictures into photoshop to create an hdr Image which is a 32bit image that has a large amount of lighting information in the picture.  I then take this hdr image of the mirrorball and unwrap it in a free program called hdrshop.  Then I can import the image into maya and use it with final gather to create an indirect illumination pass.
I Usually blur the final HDR image so it creates very soft lighting, of course if an object was very reflective I wouldnt blur it so I could get nice sharp reflections.

Compositing

Learning how to composite things for visual effects was a lot of fun.  After all this is probably the most important part of the entire process simply because it is what brings everything together to create the final result.  I used the compositing package Nuke, which was not very popular a few years ago but it has recently gained a lot of popularity.  I rendered the rodent in different passes.  each pass would contain different types of information.  Such as colour, specular, ambient occlusion, indirect lighting and fur.

This meant that when I imported them into nuke I could adjust each pass individually without effecting all the other passes.  E.g. In the final composite I was not happy with the overall look of the colour map so I used a grade node to change its colour slightly to be more pink.  I could also do things like blur the cast shadow on the floor or increase the amount of backscatter on the subsurface scattering pass.

Here is a look at the Node tree inside nuke. I like using node based compositing programs like nuke because it is very easy to organise everything.

That concludes the case study on my hairless rodent visual effects piece.  I loved every second that I worked on it and I am quite happy with how it turned out.  I hope you enjoyed reading.

I found that a huge amount of research was needed. I used google images to search for pictures of anatomy charts and pictures of elderly body-builders. I also used 3d.sk to gather a huge amount of reference. I put all these reference pictures into a folder and I would look at them constantly as I would sculpt in mudbox.

I wanted the pose to look dynamic and strong, the most important component of a fine art drawing or sculpture is the gesture, because the gesture is what determines how the muscles will look, how the skin will look and will convey the overall emotion of the piece. For instance the muscles will be stretched out or contracted depending on the pose and the skin will also change its shape around these areas. As humans age the skin looses its elasticity so there are more wrinkles and areas where the skin sags. The muscles will also be much weaker and smaller than a person 20 years younger, it’s important to keep that in the back of your mind as you sculpt because I looked at a lot of reference of people who were younger and anatomy charts and drawings of younger people and I kept having to remind myself that an elderly man wont have so much definition.

Once I had almost completed sculpting the body I imported his clothing and magic stick. It was very interesting to sculpt cloth as I have never even attempted to do that before. I think its pretty good for a first try, it was not too complicated since the amount of cloth is small and the pose is relatively simple. His magic stick did not take me too long to detail, I used one of mudbox’s stamps to get a wood texture and I did a google search for “rune symbols” and I sculpted similar markings into his stick. The object at the end of his stick is supposed to be some kind of magic crystal/meteorite.

Once I had finished sculpting everything in mudbox I then Exported it to zbrush where I made turntable movies and used its wonderful shaders to show off all the details in the pictures above. By far the hardest thing to do was to not over define everything.  Subtlety is very important when sculpting something such as this. Its very easy to get carried away and overly define all the muscles and tendons and veins etc, but it doesn’t look real and nobody’s muscles are contracted or tensed up all the time and in every pose.

I was trying to go for subtle emotions for the face. I wanted to give him the look of contempt for the thing he is cursing and give him a look like he is chanting in his head. I also wanted to try and capture the look people get in the middle of strenuous movements. I’m not sure it comes through when its a picture of the head alone though. I also wasn’t concerned about putting in such high frequency details as skin pores, the exercise wasn’t about that kind of detail, it was more about communicating the emotion of the character through his gesture.

I had tremendous fun sculpting this and I learned so much as a result. I hope you guys like it.

Below are some turntable movies of the models.

Finally I got the second version of “Sharkface” finished.  I have made a few changes to the facial details since the last WIP post and it has been shaded and textured. I used s 32bit float point displacement map as well as a normal map and the Misss skin shader was used for the subserface scattering.

Here is a look at the Colour and Specular maps that I made. Because I wanted to get as much detail as possible in the maps they were 4096×4096 in size, anything smaller just didn’t have enough detail for the UV space. The story behind Sharkface is that he’s supposed to be a mutant who lives in the sea. His skin isn’t much like human skin because it is constantly in water so he has skin similar to an amphibian but resembles the same colour and texture characteristics as a human. The little designs on his nose and neck are inspired by the markings on the Leopard shark.

I used photoshop for about 80% of the texture creation.  The other 10% was in bodypaint 3d.  There are alot of seams on this model and the textures needed to be seamless otherwise it just look badly done.  Its very easy to take the texture into bodypaint 3d and use its clone stamping tool along with some excellent brush creation tools to make the textures seamless.

I tried out a new technique for creating the specular map. Normally I would just desaturate the colour map and use Photoshop’s Dodge and Burn tools to paint in the areas with higher or lower amounts of specular value. In this case I put it in a mudbox 2009 which has texture painting capabilities in the current version, and painted the colour white at a low opacity all over the model with an alpha that looked very random and noisy. I built up areas of value in the key areas of the face that have the highest specular value such as the areas around the eyes, mouth, forehead and nose and then exported the image. This could also be done in bodypaint 3d but I wanted to try out mudbox’s new texture tools. I found I got better results using this technique for creating the specular map and I will use this technique for future texturing.

Below are the turntables of “Shark face” that were edited into my reel to replace the old sharkface, if you are intersted to see them.

Intro

The first thing I do when I am rigging is to write down everything that the character needs to be able to do. For instance I didn’t need to do an elaborate foot setup because the animation didn’t call for any complicated foot work, which is why it only has a foot roll attribute on the foot control. Its good to also write down everything that needs to be rigged so you can have a check list, It could cost a lot of time if you forget to rig a certain control. Once I have made my plan I start laying out joints, then I move onto making control objects and so on.

Here is a look at the finished rig. It took me a few days to complete and I must say that is was the most challenging rig I have ever made. There was a lot of very tricky controls that needed to be made and I didn’t go into it in the beginning knowing how I was going to do them. Because its a robot that doesn’t have any parts that deform like something organic would it meant that I didn’t have to skin it and paint weights. Which was a good thing because that would have made the rigging time so much longer.

The Rig

You will have noticed that the controls are colour coded based on their placement on the character. This is standard practice and very helpful for the animator. By having a colour coded system of controls the animator and easily differentiate between similar controls on different body parts, this means the animator won’t have to keep clicking around trying to find the controller he/she is looking in a complicated pose.

Organisation

I would say organisation is the most important feature of a rig. Things need to be set out logically and in a tidy manner so it can be easily adjusted if necessary. It also makes it far more difficult for the animator to break the rig if everything is tidy.

As is shown in the pictures of the “Outliner” and “display layers”, everything is organised. I put things like the joints, IK handles, clusters, wire deformers etc on the “DO NOT TOUCH” layer. this is simply to let the animator know that he/she doesn’t need to go fiddling with anything on that layer. I also put the controls for different body parts on different layers, this is because the animator may wish to work on one particular part at a time and doesn’t need to adjust other controls.

As is shown in the picture of the hypergraph there was alot of nodes. from geometry to joints, controls objects, constraints etc. Absolutely everything was named with a logical naming convention and then put into the appropriate groups as seen in the picture of the outliner.

The process of putting everything into groups normally takes place towards the end of the entire process because it can be impossible to know exactly what groups I am going to need as I’m going along and if you get hung up on grouping everything as you go along it could mess with some parts of the rig in ways you don’t want it to.

Features

The hand and feet controls are pretty standard. The foot setup is what’s known as the “Classic reverse foot setup”, as opposed to the group based reverse foot setup. the difference being that the classic version is based on a separate small joint hierarchy under the feet to get the “foot roll” control which is used when a character is walking. the grouped based version used a set of empty group nodes to achieve the same control and more if needed. We didn’t see the need to have many controls for the feet as we didn’t want the character to walk with the same movement as a human, because its a machine. We wanted it to have an awkward look to the way it walks as if it is imitating someone walking.

The hand controls are pretty standard also. The animator can access all the finger controls by selecting the little box that is just above the hand and then adjust them in the channel box. You can see how simple it is to get a pose by just adjusting the parameters, the animator could even select multiple attributes in the channel box and middle mouse click and drag to adjust them. It was made using maya’s “set driven key” tool.

The facial rig is rather simple also. It is a blend shape based system where the blend shapes are linked through set driven keys to the face controls. There wasn’t many blend shapes need since the character is a robot it wouldn’t have the emoting capabilities of a human. The controls are designed to give out subtle indications of emotions and because it doesn’t have a mouth it only has the eyes to convey emotion.

There are many cogs placed throughout the character. These are in place to create interesting secondary animation, since the character is a robot it needs to some kinda of complicated looking moving parts to add a more interesting look to it. It was rigged so the animator could use one control to control all cogs at once, the idea being that if the character was thinking or feeling sad the animator could adjust the speed of the cogs accordingly to create a deeper sense of emotion. The setup for the cogs was rather simple.

Procedural Animation

I used a series of very simple expressions to control each the speed of each cog with one control. For instance the line “Robot_MouthCog02_L_Geo.rotateX = time * Robot_Head_Ctrl.Cog_Speed * -2;” simply means. The rotateX attribute for the specific cog is equal to time multiplied by the main cog control speed attribute which is then multiplied by minus two. Its the minus 2 part which is where I could add a lot of randomness to the cogs. Each cog has a different number at the end of the expression so no two cogs are going at the same speed. and because this particular cog is multiplied by minus two then that means it will be going in the opposite direction to cogs where the number is positive. Video of cog animation

By far the trickiest body parts to rig were the spine and neck. I originally planned to use a “spline IK” setup with joints for the spine and neck but there were too many undesirable bi-products when using spline ik for these thing. For instance when using a spline ik for something like the spine of a character, often the joints above where the spine ik ends follow the motion of the spline ik’d bones below in an annoying fashion. I did not want this so I sought to find an alternative. after thinking and experimenting with different techniques I came up with a way to have the control I wanted without any annoying bi-products. I basically use wire deformers attached to control objects using clusters and constraints and It gave me all the control I needed.

Rigging Cables

As you can see the the spine and neck react exactly the way I intended. This setup also allowed for me to add controls in the centre of the wire if I wasn’t happy with the deformation and it was setup so the middle control would always be in the middle of the end controls. Its probably the thing I am most proud of about this rig because I thought it up from scratch and it works perfectly for what is needed.

The final tricky thing to rig was the dynamically simulated wires around the spine which attach at the hips and torso. It was simulated using maya’s new Ncloth dynamics system which proved to be quite fast and stable. It was simulated in much the same fashion as you would simulate clothing for a character. The animation was done first without any simulation then once the animation was finished I would drag the time slider the frame -100 or something around that number. which is where I would have the robots Bind pose already keyed at that frame and then It would slowly move towards its first pose the nearer the time slider got to frame 0. In other words I had to give the wires time to relax and ease into the first pose before the animation begins.

That concludes part 2 in this case study. I hope you enjoyed a brief look into the techniques used on my film and leanred something interesting in the process.

Character Design And Modelling

I will begin with a brief look at some of the concept art. There were many many different designs for the main character. It took us quite a while before we were happy with it. We were looking for something that looked androgynous, human like and could emote with only its eyes and body language. The story is set in the not too distant future so it wasn’t desirable that the robot be too advanced, we had a lot of cogs here and there on the robot. This allowed us to give it some interesting secondary animation which could give the viewer a hint on its emotional state and it also gave it an interesting “Victorian contraption” look. As it was being modelled there were several changes and tweaks made to areas around the joints. This was because parts would intersect each other as the limbs would bend which was not what we wanted.

The design was very complicated. There were many separate pieces and all together it was made up of around two hundred thousand polygons. It would be subdivided several times at render time to make it smooth.

You can see a textured and posed version of the final design for the robot character here.

The most difficult things to model were the outer shells. It can be very tricky to make grill holes in hard surfaces without getting some sort of undesirable look to its corners or edges surrounding the holes.

Character Texturing

Here is a look at some of the textures used on the robot. about 90% of the texture maps were made by my team-mate, below is the colour map for the head and limb frames. Most of the maps were 2048×2048 in size.

Designing And Modelling The Props For The Library

While we were designing the main character we were also designing the environments and props. There were so many different props that needed to be made. The idea was to make all the environments look like they were suddenly abandoned with everything left in place. The library environment had the most objects in it and all of them had to be designed. Here is an example of one of the design sheets created for the props.

Most of the props were made to be turned into subdivision surfaces at render time. This meant that I needed to use specific techniques in order to make it look right when its smoothed. This means having support edges near areas where I wanted sharp corners, just bevelling the edges was not good enough. After a prop was modelled it was then UV’d and then sent off to my team-mate for texturing, Then I would set up shaders and apply the textures to them.

Books

Because half of the film takes place in a library, the most important props were the books. A library is only really recognised as a library if it is full of books. This became a problem because we couldn’t make thousands of individual books. What we ended up doing was model 33 books and then duplicate them many times and shelve them in different configurations.

All the different book types had very similar UV layouts. This made it easy to make a lot of textures because the templates were almost all the same. Once all the books were made I then had to lay them all out on the shelves. This was very tedious, there was a lot of duplicating, moving and rotating. The books were arranged depending on the shot. If the camera was only going to see one bookcase or half a bookcase then I would only need to lay out enough books to cover the bookcases in the shot. In some of the wider shots there were several thousand books in the scene.

The floorboards were created using displacement and normal maps. This was more for experimentation, I wanted to see how far I could push the displacement capabilities while at the same time not use up too much memory. I was pretty happy with the results but if I were to do it again I would model the floorboards individually using basic box primitives with bevelled edges and just apply normal maps.

The Factory Environment

The factory environment didn’t require as many props as the library, which was good because time was running out and we couldn’t afford to make a lot of props to fill it up with. The models were very simply modelled, most if not all the models did not have many polygons. Luckily most of the objects in the factory were hard surfaced metal objects so there was no need to go overboard.

There was also a few dead robots lying around the factory. These were easier to design the the main character because they didn’t have to do anything or be on screen very long. They also needed to be very simple in design since they were to be inferior robots.

I learn’t a huge amount on this project, the best way to really learn how to do things is to dive in at the deep end and just do it. I was modelling things every day and after a little while it just became second nature and I got alot done in quite a short time.

So there it is, a brief look at some of the modelling techniques and textures used on my film. Part 02 will be a look at the rigging of the main character.

 Turntable Movie