Head Retopology

When bringing the head model from Mudbox to MAYA, I had a couple of problems. One of them was that because the model was so high poly, it was really hard to work with MAYA as it was running so slowly. The second problem was that MAYA crashed on me and I couldn’t use it for a couple of days. It came up that it was not responding and wouldn’t let me force quit the application for 2 days meaning I couldn’t even shut down my laptop.

I had my notes on retopology from the start of second semester but I also wanted to watch a tutorial on it as it had been a while since I’d seen the class tutorial. Jess told me about a short video on YouTube that went over the basics of retopology and I watched it which was very helpful.

Alongside my notes this helped me remember exactly how to do it. However, when I tried it didn’t seem like it was working quite the way I seen in class and on the tutorial.

As you can see from these images some of the lines went under the top layer of the model and made the topology look patchy. I wasn’t sure if this was due to creating the polygons too big so I tried to go smaller around the more detailed ares. However, when I started this, it didn’t look as low poly as I would’ve thought. I’m not sure whether this is normal or not. I had the surface of the model live so that the topology snapped to the surface.

Another thing I realised was that a lot of my polygons were very uneven looking and looked pretty messy which was probably due to my own fault as I don’t think I placed the points in the right place. If I had have taken my time and made the points more accurate this might have helped the shape of some of these polygons.

I used the relax tool a couple of times as I realised that this made the polygons neater and may have been relaxing them to the surface of the model, I wasn’t completely sure. When I went back on to finish off the retopology, I got it almost complete before MAYA crashed. It took ages to do but I did realise the more polygons I created the slower MAYA was getting which made it take longer than it usually would on a better laptop or computer. I may have got it completely finished if I used the Mac computers in uni.

Head Modelling

I took photographs of Jess to model for the head topology assignment. I took photos of front, left, right and front right angles to give me a better view of her facial features and shape of her head. When modelling I stuck to the basics; I did not model any hairs on the face or head. Maybe with more experimentation I would be able to do this.

To complete the assignment, I used the programme Mudbox 2017, switching between the different tools.

I started with the basic head model that was given on Mudbox and began to construct the nose. Using the sculpt tool, I stretched out the sides of the nostrils to copy the shape of Jess’. Pressing ctrl allowed me to invert the tool, which is how I created the nostrils and other indents in the skin before using the smooth tool to give a more natural skin-like finish. I found this was quite simple at the start until I started with the other features. I also experimented with the grab, pinch and flatten tools as well as the knife tool but mainly focused on the others.

I found that I completed the face features in order from easy to difficult; nose, mouth, eyes. By the time I got round to the ears, I had got the hang of the tools and the hotkeys which made it less complex.

I thought when sculpting the lips, the small space in between the top and bottom lips was the hardest to create. I couldn’t get the tool size small enough to create the small line so I had to smooth over the rough sculpt and use the grab tool to move sections of the lips and bring them together to give it a realistic look. Same goes for the eyes. It was difficult to get the crevice between the eyeball and eyelids small enough to look natural.

If I was to re-do this head model, I would spend more time familiarising myself with the tools so that I would be able to expand the use of the tools, and other settings, while modelling rather than sticking to 5/6.

Below are images of the model and the sculpture from the same 4 angles.

This slideshow requires JavaScript.

Used this tutorial to help me get the hang of some of the tools in Mudbox before using it myself:

Animating and Rendering!

Once we had the scene set in place, all that was left to do was import the files and animate then render each scene. Rachael and I met and discussed each scene, what was happening, the camera placement and how many frames per scene in correlation to the animatic. Considering that the animation was to be 24 frames per second, we used this as a guideline for the animation. We had a slight leeway with the timing as our animatic wasn’t as long as the animation should be (25 seconds max.) This was helpful as we needed some extra frames to make the animations look more natural. We divided up the scenes and split them between us to animate. There wasn’t actually a massive amount of animating to be done thankfully so we knew we’d manage.

We created a check list of each scene and then whether it had been animated or rendered so we knew what we had completed and what still had to be completed.

The scenes from the right hand side were the ones that I had to complete by animating and rendering them. I’ll insert some screen play blasts to show the un-rendered movement of each scene.

Scene 2: A close up of the apple, with the only moving camera shot. (24 frames)

Scene 4: Worm leaving its hole in the ground. (took turns as this was probably the hardest to animate)

Scene 7: Apple flying through the sky. (25 frames)

Scene 8: Apple hitting the tree. (19 frames)

Scene 9: Apple falling on the ground. (25 frames)

Scene 10: Apples falling from the tree. (24 frames)

I simply animated these scenes by using the controls on the worm and then moving the apple and also slightly moving the tree when the apple hits. Using the key frame button in the animation menu was the how I created movement.

We looked at some inspiration before animating the worm’s movements. We needed to see the different ways that worms can move and the best way was to look at other animated shorts about worms.

A Worm’s TaleRobbyEarly Hatchling Gets The Worm

When it came to rendering, we used Arnold first time round. We split the rendering across 4 of our MacBooks and a Mac Computer in uni….it takes forever! We changed the render settings:

  • Image format – tif.
  • Frame/Animation – NAME.#
  • Start/End Frame – (frames that need rendered)
  • Renderable Camera – Cam. 1/2/3 etc.
  • Width – 1600
  • Height – 900
  • Image Size – HD_720 (first), HD_1080 (final)
  • Camera (AA) – 6 (higher = takes longer)
  • and; Render Sequence once this was all set

After rendering everything with Arnold, Rachael found a renderer called FurryBall. We used this as we rendered with Arnold in HD_720, so when using FurryBall it came out much better quality for the final animation. Also, Arnold took about 6 times longer to render the frames. For the final hand in on 12th May, we plan to use Arnold to render our frames in HD_1080.

Compiling, Colour & Texture

When putting together the scene, we decided it looked a little bare with only the 2 main trees in the story so I duplicated some of the background trees that Rachael had modelled to simply fill the background. We added clouds that Jess had modelled from a previous project that didn’t come of use and also used some grass models that Rachael had done, scattering and resizing these I was able to make the scene look a little more realistic as such.

I replaced a few things in the scene and grouped everything together to tidy up the outliner box. This made it easier to see what we were using and what didn’t need to be there. There was different objects in the outliner box that weren’t in use and were unnecessary so I just deleted these. Grouping the main trees, the ground, clouds etc. By doing this it was quicker to find what we needed

Screen Shot 2017-05-11 at 15.27.31
Outliner Organisation
Screen Shot 2017-05-02 at 10.48.13
Scene (low quality render view)
Screen Shot 2017-05-11 at 15.00.25
Scene (rendered HD_1080)
Screen Shot 2017-05-11 at 15.00.10
Apple and Grass Close-Up (rendered HD_720)

Rachael and I decided that it was best to keep the models low poly and keep the colour/textures pretty plain to save render time. We needed to get it finished and we knew we had a shortage of time. To get it complete we thought this would be the best option so we knew rough time for rendering and could add more after the presentation if we wanted to. Although it looked a little plain and very cartoony, we wanted to keep the colours bright so that at least it had a happy and energetic vibe going on. We stuck to the simple lambert shader with colour and didn’t add any textures apart from the apples on the trees, we used ‘Blinn’ for these to give them a slight shine like fresh apples would be as you can see in the middle image.

We also added a skydive light and added a colour to this so that we could have a sky but also a constant light that gave similar drop off like the sun as it was an outdoor setting. We had trouble with adding shadows when transferring the file, it disappeared and we had to figure out how to get this back. We realised that the shadows were still there they just hadn’t been showing in the render view. Once we rendered in a high quality (HD_1080) with Arnold, it looked well.

Modelling and Rigging

Modelling was taking place by team members while I was taking control of fixing and finalising the animatics. As we ended up having to recreate 3 due to different criticism, this took a lot longer than expected and we had this on top of experimenting with modelling before we could start to properly put the scene together. Different objects had been modelled quick which we needed up using due to not having as much time as we would like, such as the welly boot, some simple low poly trees, basic apple, apple core experiment model. Jess created some experimental worm and orchard models and Rachael worked really hard modelling majority of the scene which I was grateful for whilst completing the animatics.

When it came to rigging, we only had to rig the worm and the welly boot. The welly was just rigged to kick forwards with one control which you can see clearly from the link:

Welly Boot Rig Video

The worm was the most difficult and we had to ask Alec for help as there was so many controls we wanted – The eyeballs, eyelids, mouth, tongue, neck, mid-body, tail and tail rotation. Alec was a great help and we ended up wth exactly what we wanted.

Worm Joints Rig Video

Worm Face Rig Video

The expressions on the worm were very important in our story as they play a majority of the worms emotions and how he reacts to certain actions and events. Simple movements such as the eyelids widening and closing to show blinking and emotions such as excitement and shock when he sees the apples.

30 Sec Animation: Pitch and Concepts

So we started by coming up with a few ideas that we could talk about to get feedback off each idea and see which would be best… to realise it was actually a presentation/pitch. We had the idea of a worm trying to get to a rotten apple core, it being kicked away and a fresh apple landing right infant of him. We liked the idea of a worm because we knew this would be easy to create in MAYA, a long with an apple, tree, grass and a foot. We tried to keep it as simple as possible. We were told that we had to give a visual display of our story or feedback on whether to continue was almost impossible.

So in the week following, we created concept sketches and models and also did some research. We originally used the Disney Pixar short ‘The Early Hatchling Gets The Worm‘. We used this and then found a couple of others after thinking of the story that we thought would be great to use for reference of how to make the worm move in MAYA: Robby and A Worm’s Tale. Both of these have different stories but also both show different movements that may be helpful to look at when trying to add movement to our worm.

FullSizeRender 9
Concepts of Main Scenes. In order: top left, bottom left, top right, bottom right

Just created these quickly with watercolours and watercolour pencils to demonstrate the main scenes and characters in the story. Other group members did some concepts as well.

I also went onto MAYA and tried to model a rotten apple core.

I tried to add some shading on different faces but just ended up looking square. So I need to look at how to blend the darker shades into the lighter. I also need to make it look a bit more rotting by making the shape a little less even and crisp looking. Arnold also wouldn’t let me render view, so I need to figure that one out!