Wednesday, 10 March 2010


  • Experimented with modelling, especially creating a human face
  • Blend shape animation for my own character
  • Using personalities to objects rather than characters with eyes and obvious features
  • Able to add the chess pieces using layers and compositing in After Effects


  • Using the Luma Key made the chess pieces seem blurred which I did not like, yet when I tried just merging the layers or adding the blending style they just looked in Xray mode
  • SHADOWS!!!! I tried helplessly for hours to create shadows, even creating the planes angled like the stairs and walkways for them to be projected onto. However, when I tried rendering the shadow layers in the different scenes my laptop consisently crashed and I had to give up to focus on compositing.
  • I could have possibly pushed the project a bit more and would have been better creating one character or even downloading a pre-rigged model so that I could concentrate on the animation alone, as this is the career path that I wish to take.

If I could do this project again...

I would watch a few more films to see if I could have found a better choice, and then looked at websites such as creative crash to find a suitable rig for me to use. I do prefer modelling them so I know the ins and outs of how they work, yet I could just make one object rather than the five or six pieces I created. This would allow me to use one more complex rig and push the possibilites of the animation with just one character. I am quite pleased with how the chess pieces move, yet think that the quality rather than quantity method pervails.

I would also allow more time at the end for rendering and find a way to get the shadows to work, as I feel this would make the pieces fit into the environment and seem part of it far more. This is my major critisism of my work and my major disappointment with it. I was also unsure of where to place lights, so a clearer scene perhaps in daylight with an area light would have been more advantageous to me.

I am quite proud of my achievement this term, I now want to take it to the next level and set clear parametres with time so that I get each section done effectively and also create a more indepth blog to coincide with everything that I am doing next term. I have learned the importance of this and not trying to fit everything in at the end of term.

Final Video

labyrinth add from Hayley Allen on Vimeo.

Maya Live

The next stage in adding to Labyrinth was Maya Live.

"Maya® LiveTM simplifies the process of matchmoving. Matchmoving is a process where you match the camera or object movement in a live-action shot with the camera in Maya. This subsequently lets you place your 3D objects into live-action footage.
Matchmoving is an iterative process where the camera movement for the Maya camera is calculated using processes of tracking and solving. Objects are tracked in the live action scene and a solver then calculates the camera position for a given image."

In Maya Live you create points in the scene that the camera tracks. After a cetain amount of points are tracked successfully Maya can solve the scene and the camera move will be created accordingly.

The tracks must be mostly green to be most successful, yellow shows some slipping but can still be used, and red is the worst and creates bad results.

I found tracking relatively simple and managed to create lots of clear green points for two of my scenes. I did however have to end up creating one movement by hand because the background colours of my Labyrinth scene were too similar and the points either had a lot of red tracking or would only track for one or two frames at a time. The images above and below show a successful track, and I will definitely try using Maya Live again on different footage in the future.

Tutorial found on the internet:


Talking about my last post, here is the image I used from Photoshop to add to my Jareth model. I gave it an effect to make it look more cartoony and not too realistic so that it fitted in with the large eyes and style of the head. I also gave it a bit more shadow on the left hand side using a new layer and a low transparency for the black paint, because I had lit the head model from one side to make him seem more mysterious. The half lit face also shows good and bad sides to the character and the confusion within Sarah's mind as to which path she follows. David Bowie's face is also half in shadow throughout my film clip selection and this also kept consistency with the real life footage.

Out of all the chess pieces, the one I struggled with the most was the Knight piece, as he is not completely round for all of him and this meant I could not create a simple CV curve and revolve it.
I therefore bought a chess set and used these photographs as image planes as reference for my chess piece:

This really helped me, and here is the Knight during the modelling stage. I used lattice points to mould nurbs spheres to how I wanted them to look and I am pleased with the results.

After modelling all of my pieces, I then had to rig them aswell. I used very simple rigs and the shapes themselves are quite simple with no additional limbs such as arms, legs etc. The only one to receive a few more joints was the Knight so that I could also make his ears move.
Here is the Queen piece with and without joints. You can also see the curve on the original piece that I used by using the image plane as a reference and tracing around one side with the curve and revolving it.

I gave my pieces wood shaders as i thought this would suit them better than say metal or trying to make them too detailed. I also chose red cherry wood for the 'evil' chess pieces because it conveys a sense of danger, and also because it would show up better on the screen than a straight black piece (you would be able to see the curves of the pieces better). The two shots below show the two different Kings to compare the shaders. I am very pleased with the results and look forward to seeing them in my scenes.

One of the main things I wanted to try during this project was to model and animate a human head, as I have not really tried it before. Earlier in my blog there are a couple of images of the opening stages of the head modelling tutorial I followed, where the head it still quite blocky and awful looking.

I carried on with the tutorial, and kept dividing the faces using the split polygon tool to make the head more detailed and create the shape that I wanted. I then smoothed the head to make the shape seem softer and more rounded.

I also added eyes so that I could make Jareth blink.

In order to make Jareth's mouth move, I created blend shapes of the head with the mouth in different positions. I created A, I, O and a normal no movement shape as these gave a variety of shape that I could key between and also I aminly needed the A and I shapes as the mouth was wide open for laughing. You can also see the controllers for his eyes in this image, for the eyes themselves, his eye lids and the pupils. I do wish that I had taken a little more time to make completely convincing facial movements such as moving around the eyebrows and the cheeks for example, yet this would have been more time consuming and I needed to concentrate on my animation.
Here is the final model. I used a photograph of David Bowie as a texture for my head model, as this easily gave me his likeness and features. I realised by adding this that the shape of the head werent quite right for him as the eyes were too wide, but in a way I quite liked this as it seems more surreal and matches the strangeness of the scene.

So here are the final Jareth animations, using David Bowie's laugh that I cut from earlier on in the film. I will add this to the end of my piece as Sarah falls, as if he has got her. I did try to add hair to the head yet struggled for hours with it not looking right, or not being the colour I wanted. In the end, I will probably use photoshop to add the hair as a seperate image so that he does not look bald in the scenes.

Original without texture:



Here was my schedule for my Industry Exercises project, having realised I have not uploaded it yet.

Week One:
  • View many films and programmes to decide on which scenes I am going to use for my project
  • Decide on a film
  • Create any concept art or begin storyboarding

Week Two:

  • Finalise my film
  • Think of what to include
  • Storyboarding final idea

Week Three:

  • Modelling
  • Begin looking into tracking the scenes

Week Four:

  • Tracking
  • Modelling
  • Texturing

Week Five:

  • Finish modelling if necessary
  • Animation

Week Six:

  • Animation

Week Seven:

  • Animation
  • Lighting

Week Eight:

  • Shadows/ Lighting
  • Rendering
  • Composite the piece together