Component 1 –

When beginning to rig a mesh, in this case it was the body of a woman, there are a few things that I had to keep in mind.

– Skeletal/bone layout (Where each bone is meant to be placed inside my mesh)

– Extension of Joints (Making sure that the joints don’t over extend)

– Skin Binding (Ensuring that there are no anomalous folds in the skin after the bind)

Skeletal/Bone Layout

The bone layout was heavily dependant on what animations I was trying to achieve from my mesh. I wanted to make sure that there was a full range of movement in the arms, legs, fingers, neck and feet so to complete the animation I desired.

I began with the arm bones, where I inputted the shoulder bone as my starting point. From here I created more joints to follow the structure of a human skeleton, however, I incorporated a ‘roll’ bone between the elbow and the wrist. This allowed there to be a slight orientation of the forearm ‘muscles’ when the wrist rotated. This created a higher level of realism for my mesh as it acted as though the ‘extensor’ muscles were being activated in my object’s body.

Each joint I created needed to be appropriately orientated on its pivot point to allow for a realistic rotation of its origin. For example, my elbow bone needed to be edited so that the axis followed that of the shoulder bone, so that if there was an alteration in its rotation, the wrist and other bones below it in the hierarchy would also follow.

After correct implementation of the elbow and wrist bones, the next thing I set out were the finger bones. This was very time consuming as each orientation of each knuckle and joint needed to be correct otherwise my control systems would not work to their full potential.

Not only did I use the above image as reference but also used my own hand as a live representation of what should happen with the finger movement/orientation.

This was highly useful as being able to refer to something in real life is a lot better than to follow guides through a picture.

After completion of the arms and hands I began to shape the head bones, starting from the neck as this would connect well to the top of my subject’s spine.

I used one bone for the scapula, one for the vertebrae, another for the back of the jaw which also extended to the top of the skull, and then a final bone coming towards the front of the jaw. This ensured I could get a full range of movement in my subject’s head without any problems.

I then moved towards the lower half of the body, this was relatively easy due to the lack of fingers and bones from the waist down.

I used a bone for the hip, one down to the knee, then to the ankle, one for the ball of the foot and one of the toe tip. There was no need for individual toe bones as I was not intending on animating them or setting them up for rigging.

Extension Of Joints –

It was vital for bones such as the knee and elbow that they could not have hyperextension (extend further than what they care capable of). I made sure that this did not happen by having control systems linked to both the knees and elbows that ensures of correct rotation and no inverting when moving the rig around. The image below shows a representation of what hyperextension of the knee is.

However, there also needed to be limits on the finger rig, so that the control systems could also not overextend the fingers resulting in a lack of realism in the rig.

The right of this picture shows what the furthermost extension of the fingers can reach to, anything beyond this would look unnatural on the joints on my subject.

Skin Binding –

As can be seen in the picture below, if an arm is out and away from the subject, there is a pull of the outer chest and underarm of the skin, however, the opposite side shows that this stretched skin comes back and almost folds upon itself once the arm is down by the woman’s side.

This was vital when it came to weight painting my character, I needed to make sure that the areas that are affected by each joint were almost perfect so to replicate that of normal skin and increase the reality of my rig.

Another area that was difficult to weight paint was the rear of the knee, after many attempts and editing I was able to create a fold in the rear of the knee when my character bent it. It was important to make sure that the skin of my subject also stretched back to its origin after the bend in the knee had straightened, I made sure of this by set keying animations and moving the skin manually as this was not able to stretch back properly after the bend on its own.

Pre-Prodution

My character, Ryan, is a driven 15 year old and has been brought up in the southern state of New Mexico. From a young age he had been brought up around the world of athletics by his mother, Jena, who had previously held a record in the 1968 Olympics for the fastest 200m hurdler.

Ryan had been training since a young boy by his mother so maintains a good level of fitness and his body shape shows this as he showcases a lean 60kg of muscle.

His aspirations are to become the next Olympian in their family tree, however, no mater how hard he tried or how many hours he put in day and night, he could never perfect his technique.

Besides his problems with his hurdling he continues to be a proud boy and so keeps his shoulders back and his head up high.

This contrasts with his running technique which represents his determination, this is shown by his head and body staying low to allow for a more streamline approach to the jump.

He continues to hold a positive mindset to the sport and as he believes that one day he will be able to replicate the technique of his mother and become an Olympian, he never slumps his shoulders or becomes angry with himself, he uses every attempt as a step forward.

To represent Ryan in my animation I have chosen to implement the Mixamo character ‘Bryce’ as he represents a young looking boy who could range from anywhere between 14-18 years of age which is very suitable to my character, Ryan.

The character, Bryce, shows good posture and a good physique for fitness and exercise which is perfect for Ryan due to the training hours he puts in with his mother.

Furthermore, he is seen to be wearing shorts and a tight top, this is also very fitting for my character as hurdlers wear skin tight uniforms that do not go past knee length.

The trainers that Bryce is wearing will also positively contribute to the appearance of Ryan as due to the exercise he will need to be wearing something suitable for the sport.

Vicon Blade Preperation –

To create the starting point for my Motion Capture I used a MOCAP suit and placed the Capture points appropriately on the suit to allow for the MOCAP camera setup to detect each point on the suit and map out the movements of my subject.

I then set up the cameras ready to record my subject. After setup I told my subject to complete tasks that I wanted to be a part of my animation.

The first animation shows my subject in a set up similar to the start of a race on a track. My subject then sets off out of the blocks and begins to run.

The second animation is my subject continuing to run, the run is steady as he is building up to the beginning of the jump sequence.

The third animation shows my subject getting into a more streamline position and attempting the jump. However, I told my subject to complete the jump with poor technique to replicate the problems that my character, Ryan, faces.

The final animation shows my subject continuing to run and finishing the race. Beside the poor effort for hurdling it shows my subject ina happy mood as this portrays Ryan always being in a positive mood and seeing every attempt as a success in moving forward to perfecting his technique.

After recording I opened the files in Vicon and reconstructed them to begin labelling each point that the cameras had captured for each animation.

I detected the labelling cluster on the back of my subject as this helps the software detect the front and back of my animation. 

I started with the head and worked my way to the feet as this was following the pipeline of the labelling set up.

After labelling all of my capture points I then created a solve which implements a skeleton into my labelled capture points, this skeleton then follows my labels and allows the animation to run through but with a body for better reference.

after the solve had completed I saved my scene as an edited file and then I exported my edited scene as an .fbx file, this is useful as then I can take this into different softwares and furthermore complete my animation with my character included.

 

 

For this component we were asked to record a scene using an animated camera pan movement whether that be vertical, horizontal or both. We then were asked to create an object using Maya and import it into the scene whilst making it look as realistic as possible using 3D tracking and colour grades to ‘attach’ the object to the scene we had filmed.

Filming My Scene –

I started this project by brainstorming for places that would be appropriate to use in tracking, and also somewhere that would be suitable for attaching my object to.

I then decided to film my video on the piece of land outside of my building as it had a large open space where I could implement my object, and had minimal intereactions with the floor so it could be useful when tracking.

Creating my Object –

Once I was happy with the film I had recorded, I then began to think about what object I could create that would fit well in the scene and could look very realistic. I began with the idea of a jet that would fly through the air just above the buildings in the background, I constructed it in Maya, textured and rendered it out to begin implementing it into nuke. However, after considering that the task was to extend the scene rather than have a moving object in the background I decided against the idea and began to brainstorm again.

I then decided that a fountain would dit very well in the middle of the grassy area in thge middle of my shot, It could be colour corrected to fit that of the buildings and use stone texturing to further increase its realisticality in the scene.

I tried to ensure that the fountain did not look contemporary or minimalistic as i wanted to make sure that it didnt stand out like a sore thumb in my scene as this would reduce the likeness of it being there.

Once I was happy with the model shape, I began to texture the fountain using an image for the colour. Again making sure that it could fit in the scene so I decided to use a stone texture.

I increased the specular roughness of the object to match the properties of real stone.

Tracking My Video In Nuke –

Once in Nuke I used the CameraTracker node to render through my footage and hold points that would follow through to the end of the clip. I then needed to get rid of each point that would cause a problem in my scene.

I deleted all sky points as it could lock onto clouds and they move which would interact with my tracking, and then I also reduced the Max Error in the CameraTracker node to ensure that any red coloured tracks would be excluded from my scene.

After this I created my scene which also created my pointcloud node, here I can switch to 3d view and see each track in my 3d space. It helped me to visualise where my scene was and how to incorporate my object. 

After creating my track I took the camera properties into Maya, where I applied these setting back in so I could see exactly what I would see in Nuke. This made it much easier to implement my object as I could see the track in maya and allign my object to those points.

I scaled my fountaion down and fit it to the footage accordingly and then began to create my shadow for the floor.

I created my ground plane to sit on the bottom of my fountain as it was going to be on the floor. I applied the material ‘aiShadowMatte’ which takes away the visibility of that plane in render view but also allows for the shadows to be casted onto that plane.

I then needed to create a lighting set up that would match that of the scene, this took a while as I needed to carefully adjust each area light to fit the scene. One thing that I had noticed once applying the lights was that the shadow had become a lot darker than those of the objects in my original footage.

This was because the standard settings for the aiShadowMatte were active…

I then modified these until I was happy with the shadow I could see from my fountain in the scene.

This made sure that the shadow was not dark enough to look black on the grassy area around the object and brought down the opacity from other areas of light in the original footage.

After being happy with how my created object looked in my scene I rendered out the passes to I could then take them in to nuke and apply them separately.

I needed to make sure that I applied ‘Ambient Occlusion’ and my ‘Shadowmatte’ layers separately and they would come out as independant renders.

Nuke Structure –

Afgter exporting my maya footage and renders of each pass, I then took those files back into Nuke so I could then merge them all together again and start to apply colour corrections and grades where I thought fit.

After playing around with different colour saturation movements and gain alterations, I was happy with my fountain object and was ready to implement it back into my scene…

making sure that my nodes were set up this way ensured that I could see my fountain over the top of my footage, and so I was ready to render out what I had put together.

Above is what my footage looked like without colour corrections and alterations.

And below is what my footage looked like after my improvements.

Even though there is not a hhuge change colour wise, the object in the swecind video can be seen to look more fitting than that of the first video.

And finally here is the breakdown of how I produced my final render.

 

Task 1 – Spiderman

I began this task by setting out my background plate in Nuke, in this case it was the footage of Spiderman running up to a wall and doing a flip inside what looks to be a gymnasium.

I set up the background with a keylight using the Spiderman plate as the ‘source’. I then included a Roto node, and connected the ‘bg’ pipe to the Spiderman plate as this is the one I was wanting to modify, whilst also connecting the ‘OutM’ pipe from the roto to the Keylight. 

After setting this up, I then selected my Roto node and used the bezier tool to outline the body of Spiderman into different parts. I started with the head, and worked my way through each frame to edit the points of the roto to match those of my subject. Staying precise to theb point became difficult due to heavy blur in the original footage, however, using the feather setting on the roto helped maintain the quality throughout.

After finishing the roto, I checked my progress by changing the keylight settings to ‘inverted alpha’ this allowed me to see exaclty what was happening inside the roto and give me a clear picture to how my roto work had progressed.

After this, I thought it was best to connect the background that I was wanting to implement my subject onto, as seeing them together would help me notice any problems with lighting between the two and scale. To do so I needed to reformat my background image as it was different to that of the Spiderman’s.

Following my keylight, I added a motion blur to my subject, as doing so created a more visually pleasing output. I then also used a ColourCorrect node to slightly edit my Spiderman to look more custom to the scene. I lowered my contrast and saturation as I thought they were too bold for my chosen background and it’s lighting. and also transformed the position of my subject to make sure it looked appropriate regarding the background.

I also animated the position of my subject throughout my footage, as when played back, it was noticed that Spiderman’s foot had broken the perspective and looked like it had gone inside of the pavement in my background. I fixed this by translating my subject up to when he takes a step, this gave the illusion that he had gained altitude whilst running onto the pavement which helped the footage look more realistic.

After watching it back I decided that I wanted to add a shadow to increase the reality of my video, and the easiest way to do this was to replicate the roto that I had used for the original Spiderman plate, and skew and translate it. I needed to make sure that whenever the original Spiderman’s feet touched the ground, the feet from the shadow of Spiderman came into contact with the original.

This however became a problem when the Spiderman flips off from the wall, as the shadow distorts from the perpective and ruins the whole look of my video. This meant that I had to animate the translation of my shadow to fit the scene. I animated the skew settings, the rotate and the transform attributes to make sure that the end result replicated that of reality.

My finished node tree was neat and tidy with the use of dot nodes to ensure the nodes did not become tangled or overlapped.

 

Task 2 – SkyDiver

As with the Spiderman task, I began by inserting my footage of the Skydiver, in this case he was on a box in a blue screen set up.

I then started my roto just as I did with the Spiderman project. However, instead of going through frame by frame to change the points of the roto; this time I created a roto for each different shade of blue thast there was on the background.

I did this as I was intending to use a keylight to select that blue colour and get rid of it, meaning that I would be left with my subject.

Here is the set up of my keylight and rotos that came from the Skydiver footage.

As explained, each of these roto/keylight’ s contains a different colour of blue that I am wanting to erase from the background.

I selected these shades of blue by going into the keylight node, selecting the colour picker and clicking ‘ctrl’+’shift’ to create the target to where I was wanting to select. I then could check how well each keylight was working by changing its OutM Component to inverted alpha, when changed, it shows a higher contrast between the colour that is being taken away (black) and the colours that are left intact (white).

At points where my rotos were showing too much grey, meaning that is was a poorly executed keylight, I changed this by upping the screen gain, so that the greys would fade back to black. The only problem with this is that often I lost some detail if I pushed the gain too far.

After making sure each of my keylights were working correctly with my colourpicker selections, I combined them together using an AddMix node. Making sure thast for each of these nodes I had inserted, I had set them as pipe ‘b’ so that they would add over the pipe ‘a’.

Once this had been fully set up and structured using the dot nodes, I then incorporated my next node which was a camera shake. This helped to ensure that my video seemed as thought it had been filmed by another Skydiver that had taken the footage whilst following the first subject. My next addition was the plate that contained the picture of the ground, I had animated this using the transform node as I wanted to follow the impression that the Skydiver was falling so I had to increase the scale very slightly throughout the video.

I then wanted to create different layers on the clouds so I went back up the pipeline to put them in where I wanted them situated.  Giving them differernt ColourCorrect properties and giving one layer a camera shake really improved the reality of my video as the more layers that I had added came the better the illusion of realism.

I made sure that there were different settings for each camera shake, where the clouds that wre closest to the SkyDiver had the larger intensity of shake compared to my Ground plate that had the least amount of intensity. This helped to improve the perspective within my footage with greatened the effect of the video being real.

Logo Design –

Starting the process of designing a logo, I thought about what brands have the best logos and what do they include? After searching around and looking into these questions I found that most brands that have a successful logo, also have an animal of some sort to represent themselves.

One of the best companies to use an animal for their logo is PUMA.

However, there are many other logos that are also successful :

I began to think of names that would suit what I wanted to represent in the logo. For the logo I had made the decision that I wanted to create a panda and have the word ‘Muddy’ in the name. After thinking about what flows well, I came to the brand name ‘Muddy Panda Media’.

After the decision of the name I began to start the design process of my logo.

Below are some of the first designs I had to fit the brand name.

However, after getting feedback on how my logos were looking, I found out that they were starting to become more of an illustration rather than a logo, and that to correct this I would need to simplify the logo massively and even remove the colours to maintain a level of simplicity throughout.

Due to the logo needing less colour, I decided that ‘Muddy’ had not been the best word to go with and it would be hard to show mud with just black and white colours. Prior to this I decided that ‘Soggy’ would be a better word to go for as it would be easier to create a teardrop effect and it be noticeable rather than a black patch of mud.

Below are the designs I had to suit the change of the brand name.

It is clear that the colours have been taken down to just black and white, and the original illustrations have been further simplified to where the panda now doesn’t have any body, legs, arms, or even eyes. However, even though these logo ideas do not have eyes it can be represented by the black patches that stand in the place of the eyes.

After considering what would look better between the two and having feedback on which comes across as more professional, the logo below is what I ended up using for the final piece.

Character Design –

Just like with the logos design, I started to think about which characters are most memorable.

The characters below are ones that have been popular over years and have the ne impact.

After looking through these character, I decided that I wanted create one that advertised itself of a cereal box.

I began with this design for the front and side views

I liked the idea of having the hippo wearing the hoop has it incorporates the cereal shape into the character and makes it more memorable.

After these two designs I began on the pose design. This one I wanted to make look appealing and happy as the audience for this cereal is kids.

I included a smile, and a wave in the air with both arms, to look welcoming and happy. Then started to take the hoop and put it on an angle for where the body would move it.

 

 

 

Making Improvements To My Asset –

To show further development to my asset, I decided to remove the original barrel of the gun that I had created in the first model, and replaced this with an improved barrel that took more detail. There were more objects used in the second barrel whilst also using the boolean tool to create holes within it to increase the level of detail.

 

The next thing I set to improve was the sights on the rifle. The original one did not seem to be very realistic and looked largely out of proportion to the rest of the weapon, due to this I removed the original sights and replaced it with one that I thought was a better fit to the gun’s look.

In the picture above it is noticeable that I also included a new material. This material was created to replicate that of a very translucent and shiny surface. I did this by using the ‘blinn’ material and increasing the transparency property to a higher value, whilst also increasing the specular roll off for an extra shine. I lowered the Eccentricity down to a value of 0 to make sure that the reflection that came from the material was sharp.

The last thing I did to extend this model, was to create a better lighting, one that was not as harsh and directional as seen in the first project. To do this, I set up an arnold skydome. This was great to producing a light that covered all parts of the weapon. I set the intensity up high to bring out the colours of the materials that I had selected in the Hypershader, especially the shiny surface of the ‘blinn’ material I had set for the dark grey metal parts of the weapon. However, after adjusting the skydome, I had no bottom surface to cast a subtle shadow on. After creating a floor wiht rounded corners to create a forever effect, I then had to adjust the skydome lights again to make sure that the shadow that was being casted onto the fllow wasnt as sharp as the previous one.

In the end I really dimmed the intensity down to a point where the shadows are only just noticable to give the effect that there is a floor and not just a grey backdrop.

I think that the lighting in the current project is a large improvement on the first project as the contrast between the darker parts of the gun and the brightness of the sight’s glass-like material add a huge amount of contrast to the final render, increasing its ability to look like a real-life object.

In this challenge we were given a freedom of what we wanted to create. We had the choice of a Character/Creature, a costume, or a vehicle.

I had chosen a costume and wanted to make it give a character stealth, agent feeling whilst also incorporating colours and items that would not be seen on earth as I wanted my costume to have a space or other-world feeling to it.

I did this by using basic clothing examples, a bomber jacket, jeans, and black boots.

I then started to think about what colour you don’t normally see on a normal item of clothing, I went with a bright cyan colour. This also worked well as it contrasted with the black and made it look as though it was a light, or neon shine on the clothes.

As my costume was for a character of stealth, I included a belt going from the top left shoulder to the bottom right, this was hidden under the bomber jacket and can only be seen in the opening of the jacket zip.

To add to the unearthly feeling of the costume, as I had only added a strange colour so far, I began to think about how the character would be able to breathe on another planet if it had come from Earth. This is when I started to add a mask and goggles, I looked at other characters and other goggles to see the style of them, below are taken from the character ‘Cyclops’ from X-Men –

I started with what seemed to be glasses and a small mask that only covered the mouth. However, I then built on this and moulded the glasses to a set of large goggles with a headstrap, and a larger mask tat covered most of the jaw. Although, as I wanted this headwear to be left to a minimum, I decided that the goggles and mask should not be joined and they should be left separate.

After putting these together I came to the result of my ideas for the costume.

Below is my finished idea for a costume that would be used for a stealthy space agent who had come from Earth.

 

 

This week we had to create our own vehicle, the guideline were given to us on what they were to include, however the rest of the vehicle was for our own creation.

Vehicle 1 Guidelines:

The design must be streamlined for both space and in a planetary atmosphere, wings are optional but should not be too large.

It will not be a military ship so there are no guns or heavy armour allowed.

Vehicle 2 Guidelines:

No larger than a military tank.

Must have exactly 6 wheel or 2 tank-like tracks.

It should be visually interesting but also be practical.

Not a military vehicle so no guns or heavy armour allowed.

This week’s challenge was to create a hybrid of a human and an animal of our choice. Below you can see the transformations between the animal and human from different angles and how they are moulded.

Parrot – 

Fish – 

Rhino – 

Wasp – 

The second part of the task was then to take the hybrids that we had created, and give them two different personalities. This can be seen in the image below, where each hybrid has a male and female version.

Top left – Fish

Top Right – Rhino

Bottom left – Wasp

Bottom Right – Parrot

Bevel –

The bevel tool is used to create more of a rounded look rather than one harsh line to end at the edge of an object. 

Boolean –

The boolean tool allows you to select one shape, highlight another, and extract the second shape away from the first. 

In the example below, I have used the ‘difference’ boolean to extract a cylinder from a cube. 

Inserting Edge Loops –

Edge loops are a way of creating an edge within a shape. 

In the example below, I have selected the edge loop and placed it along the top horizontal line of the cube to create an edge that you can see travels vertically down the cube.

Extrude –

Extrude can be used to create more to the shape you started with. 

The extrude can be enlarged, moved up, down, left and right, and when combining this with another extrude, you are able to create more and more shapes that are connected to the original.

Below you can see that I had started with a cube, extruded and made the extrude plate smaller, extruded again, and then moved the object to the right to then make a smaller version of the original cube.

HyperShade and Texturing –

Using HyperShade, you are able to create materials and a new look for the object you are using.

I started with a normal cube.

Opened the HyperShade window

Created a new material that I wanted to assign to my object. Selected ‘file’ as I wanted to use a picture from my library.

Selected the picture from the file’s destination.

And then placed the image file into to the material. After this I assigned the material back to the object.

Lighting –

If you turn on the all lights button in Maya, the view will look like this, this is because there are no lights pr no lights have been created yet. 

Using a spotlight is a good way to show lights and shadows in the object. In the attribute editor you can change these settings for the spotlight and make it more intense, have a bigger dropoff etc. 

Once in the render view, you will be able to see a real-time shot of what the object looks like through the camera selected. 

I used the skills above to make my own model. I used reference imagery and created my own lights to make the model look as real as possible and have a professional lighting.

 

SA80 Rifle –

 

SA80 Rifle (Image)  –

SA80 Rifle (My rendered model) –