In order to prepare for the lesson on UV mapping, I did a short exercise myself experimenting with planar mapping using a rainbow target I quickly made in Photoshop.
During the lesson itself we were taught more basics of UV mapping, using a model made by the tutor.
I felt like I needed to learn more, so I did some practice by myself after the lesson. I modelled a fish quickly in Maya and painted a texture map in Photoshop to go with it. Although the model and colours are a little rushed and basic, it definitely benefitted my learning.
During the lesson we got back into hang of modelling by making teapots in Maya.
After making a number of different teapots our assignment was to give our teapot a character. I wanted to avoid the Beauty and the Beast cliché of just giving it a mouth and eyes, so I tried to approach it in a different way by using form and shape.
Following on from the character design sketches, I grew attached to the lizard designs. As I have a bearded dragon of my own I thought I'd make the design more simplistic and realistic. I didn't feel the need to add anything extra to his anatomy as he has plenty of character as it is! I started off by doing some life drawing of him in his vivarium.
In one of our lessons we tried making clay sculptures of our designs to get an idea of what our character would look like in 3D; unfortunately I missed this lesson due to ill health. I tried to catch up in my own time, however I missed out some of the fundamentals of modelling with clay - something I've never actually done before - such as making an armature, or buying better quality clay!
It's suffice to say that I should probably stick with digital modelling...!
Here is a collection of some of the character design sketches I accumulated over the course of two weeks. I started off with humanistic characters but after a while I wanted to try something different, so I continued with animalistic designs which I think were a little more successful.
I then started fine tuning some of the designs, focusing on a little elephant, 'Ellie', and a mutant lizard/rat/hamster...
The focus for this term will be centered around character design, resulting in a fully modelled character by the end of the project! This will lead on to next term's project - environmental design - so when thinking of characters we need to keep story and location in mind. We started by looking at other examples of character design and what makes a distinct, successful design.
We watched the animation, 'French Roast' for inspiration - all of the characters were unique and memorable, and importantly, all have distinctive silhouettes. As an exercise, we were told to try doing our own designs for the characters. This was fun, but challenging considering I already had the memorable designs from the animation in my mind.
After a long summer with not much drawing practice at all, I thought it would be a good idea to get back in the habit of drawing before I attempted some proper character design. I did this by practicing a little life drawing and doing quick sketches of friends and classmates.
We were told to start thinking about environments for our characters and to come up with some preliminary ideas in the form of quick sketches and thumbnails. We had to come up with three environments with a story arch in mind:
1. Home - Where the character lives, feels comfortable. The beginning of the story.
2. Somewhere new - A new area the character has to explore. Tension rising.
3. Somewhere scary - An alien location where the drama happens. Dramatic contrast.
I hadn't thought of a strong story yet, so it was a good exercise to get me to start thinking about the adventures my character, or my bearded dragon, could have. Instead of putting him where he would normally be in the wild - the desert, large areas of sand and rocks - I decided I could put him somewhere similar to where he lives now - a manmade vivarium in a flat in London. I decided the latter was not necessarily as visually rich, but as a modeller, it would give me more challenges and a lot more to work with.
I thought it would make it more visually exciting to see the environment through the lizards eyes, so instead of a small plastic plant climbing up the side of the vivarium wall, it could be a huge burst of fauna, like vines in a jungle, or instead of a food bowl with a small pile of rocket, it could be a giant mountain of food, through his perspective. I imagined that he could climb up this massive tree trunk (which in reality is a medium sized piece of driftwood) and find a way out through the back of his little world (vivarium) into the next environment. In order for him to make this unlikely move, I could introduce another character, a cricket, to lead him there.
There he meets a swarm of cables and electricals which he has to travel down in order to try to catch the cricket. He has to face many man made objects, such as a colossal lamp and a blaring television. The cricket starts climbing up a tall, toppling pile of books and magazines, leading the lizard to his third environment...
Which takes him to a windowsill with the window slightly open. He looks over the edge to see something far more frightening than the cables/lamp/TV - a gigantic drop and a busy road underneath. Terrified, he takes a few steps back and falls over the edge of the windowsill and on to the floor again. He turns around and sees the cricket standing outside his vivarium, beckoning him back home.
The story isn't well developed or imaginative, but I wanted it to be simple and environment driven, as opposed to action driven. I'm sure I will probably end up with something completely different by next term, but it serves as a good starting point. The last frame shows a simple floor plan.
In my free time over the past couple of weeks I have been following another modelling tutorial that has no connection to character design, but was good practice for modelling in Maya. The tutorial was extremely helpful in that it took me through the many different ways of modelling in Maya, however if I was modelling a pen knife of my own I probably would have done it in a slightly different, simpler way!
After a successful formative assessment with my tutor, I devoted a few days to do some research into what makes a successful, humanistic animation. I usually draw inspiration from one of my favourite 2D animators, Glen Keane, due to his subtle, realistic animation, however this time round I thought it'd make more sense to take reference from 3D animation.
I
re-watched Toy Story, the first feature-length animated film made by the
pioneers behind Pixar Studios, which provided valuable influence. What was helpful to see is that it's still possible to create realistic animation whilst exaggerating key poses, and Pixar have managed to do it in a comical way without being pantomime-like.
I also used the archives from http://www.11secondclub.com/ to look at more hand gestures and techniques in animating. For example, I looked at this humorous piece which again, exaggerates key poses without looking entirely unrealistic:
As
I was animating a human rig, I wanted it to appear as human-like as
possible, within the boundaries of my skill. As well as filming myself
act out the dialogue (see previous post), I watched some extensive
videos on body language, looking at common gestures and the subtlety of
human movement.
After the research stage, I thought it would be appropriate to create a gantt chart for the upcoming 5 weeks - there was a huge amount to do and I had to factor in a trip to Barcelona to go to the 'OFFF' visual arts & animation festival. The event itself was invaluable in terms of inspiration, however it meant that I found it very difficult to stick to my schedule!
Before I began blocking, however, I wanted to study my acting reference a little more. When I drew flow lines and wrote down precise movement analysis for my walk cycle reference during my last project, it proved to be extremely useful and definitely benefitted my final animation in terms of realism. Here's hoping I can do the same again!
When I began blocking the key poses, the things I was looking out for were flow lines, silhouettes and lines of action. These are things I pay very close attention to in life drawing classes, so my time spent drawing and anaylsing the human body has certainly paid off. Mainly, it meant that I could craft some really emphatic poses, which would hopefully tie my animation together in making it look like a proper performance.
Examples in posing.
I like the bottom right image especially - the angles in his torso create a strong line of action which is followed by the neck and head, and the arms & hands show good juxtaposition which provides visual interest.
Taking reference from an image I took of my own hands - very difficult to pose and animate correctly!
Eventually, I was able to get past the blocking stage of animation, resulting in this playblast:
I then spent a long, painstaking time working on inbetweens, which I have to admit was my least favourite part of the entire project! I did this until I had keyed most of the frames, not all, and then turned to the graph editor to change the tangents from stepped to plateau - I found this to be more useful than spline as it created lots of flat tangents too, which I was in need of. I did this for each body part separately so as not to be overwhelmed by numerous joints rotating out of control. I did so in this order:
Torso
Neck
Head
Shoulders
Elbows
Wrists
Hands & Fingers
Thighs
Face - Jaw
- Lips
- Teeth
- Eyebrows
- Nose & Cheeks
- Eyes & Lids
For each body part, I changed the tangents, adjusted the curves in the graph editor to create smooth lines, anticipations, ease ins and outs, offsets, and define weight in general.
Graph Editor: Anticipation
Adjusting keys for the mouth one frame faster, for more effective lip sync
(as humans see before they hear)
Evidence of saccadic eye movement.
(The eye scanning/moving from A-B-C)
Realistic eye animation. The pupil tends to follow and stay 1/3 behind the upper eye-lid.
(Also used techniques such as half blinks to show thought - closing lids in 2 frames, then opening in 3, as opposed to a usual blink of closing in 2 frames, staying closed for 2, and opening in 4)
Graph clean up for just one finger!
Using 'physical sun & sky' as a light source, which streamed through a glass texture.
Despite what seemed to be effective planning, I began to run out of time towards the end of term. This was due to a number of unavoidable circumstances, such as ill health and a week in Spain, however I did the best I could in the time that I had. My original intention was to animate my female character, but I was only able to give her a tiny bit of movement in the end. This meant that I had to change my shot sequence & edit, however I actually prefer the end result of just two different shots to establish what was my main character's imagination, and what was actually happening in the room.
Using playblasts I tried 2 different editing techniques, both with different shots, different framing, and different lenses/apertures as opposed to using Maya's regular 30mm. Unfortunately, due to time constraints I am not able to upload my tests! I've decided to choose the final edit as I feel it's the most effective comically - fortunately a few friends and family members of mine agree!
I came across numerous problems during the project - such as rendering and lighting. Although I was able to focus a large amount of time making my character as humanistic and life like as possible, I still wanted him to look good in the scene, so I feel I may have wasted time with choosing correct lighting. As I was using 'physical sun & sky', my render time was almost 2 minutes per frame, so I had to give this up very quickly!
Eventually, I managed to get the render time down to 15 seconds per frame whilst still having effective lighting. During this summer I plan to work more on the piece, perhaps refining and exaggerating the lip sync to make the piece more emphatic. I also would ike to re-animate the hands using constraints, as the movement appears quite jagged. I could also experiment with different posing, of course, such as getting the rig to stand up during points of high energy - I think this might make it more effective overall.
Finally, I would like to change the sequence settings before uploading it online, as the online conversion has made the audio slightly out of sync, and frustratingly, made the arm gestures quite jerky as opposed to its smooth appearance on my laptop.
I think I've accomplished a huge amount this term, but I don't want the the learning to end now that the holidays are here! I plan on refining this final piece, and going on to learn new animation techniques and to take my modelling further.
Below are notes I made in preparation for this unit's performance piece, looking at the who/what/when/where/how of the scene, the considerations I've made during the planning process, a lay-out for the scene location, and thoughts behind the appearance of my main character.
(click to enlarge)
A revised set-design.
Although we will not be marked on anything other than the performance animation, I hope to have my backdrop as close to (my idea of) a psychologist's office as possible, in order to set the correct tone of the piece. The fact that my dialogue is essentially a daydream could be confusing enough without having to figure out where, and therefore who, the characters are.
My finished storyboard.
This was certainly a challenging storyboard to make! Unlike anything else I've done, all of the action takes place in one small, concentrated area. I tried not to rely too heavily on my acting performance, and drew hand gestures that seemed relevant to the text, including the subtext behind it. As the dialogue is mostly Dr. Sobel's daydream, I didn't want to draw any focus away from him, which is why I've only used three different shots.
Lots of arm movement!
Below is my animatic. Spot the 'W' pose! Although this is seen as quite a clichéd, pantomime-like movement for performance animation, I believe it compliments the dialogue and would be appropriate for the frustration and disbelief my character is feeling.
Although the 'Morpheus' rig has excellent customisation capabilities, I couldn't quite find what I wanted from the pre-set features. To get around this, I took the original textures into Photoshop and spent a couple hours going back and forth into Maya, tweaking them until they looked right (I did not realise initially which areas of the texture would be stretched in order to wrap around the model's head).
I figured that as Dr. Ben Sobel is a tired, frustrated man with little interest in his job, it should definitely reflect in his appearance! I decided to make his skin tone quite washed out, implying that he doesn't get outdoors much, grey lines under his eyes to indicate tiredness, and a 5 o'clock shadow (painted with a 1px brush!) to portray lack of care in his appearance. I also painted over one of the eye textures to make them look grey and lifeless.
With Caroline, I wanted to create his opposite - tanned skin, bright, blue eyes and makeup-
implying care and thought in her appearance, not to mention youth. In
the film, Caroline is portrayed as a middle-aged woman, but I feel that
for the sake of my scene, the audience needs to relate to the main
character, and therefore feel resentment for Caroline. Although
blonde-hair-blue-eyes could be seen as a rather clichéd approach to a young, perhaps
materialistic, woman who moans about her failing love life, it will at
least provide an interesting, contrasting aesthetic to the scene.
The first thing I wanted to do in preparation for my animation was tackling the dialogue. I converted my chosen scene from 'Analyze This' into an mp3, and listened to it on repeatin hope that it might find its way into my subconscious. I then wrote out the text to properly study the nuances in the speech.
I did my best to notate the essentials to gain a true understanding of the text and dialogue. I based most of this on the sound recording of Billy Crystal's voice, as opposed to my own, as that is what I'll be animating. By this stage I was definitely able to recite the text in my sleep, so despite all embarrassment, I decided to film myself as a reference for my animation. I took many, many takes of myself acting out the piece, and I found my performances were becoming more confident towards the final takes. As I went along, I experimented with different actions and movement, as opposed to trying to do the same thing again and again. This meant that I could have a wide breadth of choice when it came to choosing which parts I wanted to reference.
Above is a compilation of a number of different takes, split up into sections to make it easier to find which reference to work from. It's clear that I'm not exactly comfortable in front of the camera and my semi-American accent only makes things worse - despite the fact that a British accent could have made a subtle change the dialogue and facial movements. Also, it's obvious that I was too embarrassed to shout fully at the end, which could have affected my acting; something that should be rectified if I am to film myself performing this dialogue again. For an example of one full take, see: http://www.youtube.com/watch?feature=endscreen&NR=1&v=s2bJki5gAgs
I chose to remain seated for the scene for a number of reasons: - I feel it would be unnatural for a psychologist to stand up mid-session, daydreaming or otherwise. - I don't feel that standing up would give the dialogue any extra emphasis or meaning. - I want to concentrate on animating a realistic performance, and as most of the expression lies in the human face, shoulders and hands, I don't want to give my self extra work by getting my character to stand up and walk around. - I'd have to compromise the shot I'd want in order to show him standing, which would distract from the main focus (face, shoulders, hands).
Only two days after filming my acting reference, we were scheduled an acting lesson from Sarah Perry, a professional acting coach - if only I knew this before I attempted filming myself! In this session she took us through two main acting approaches - the Stanislavski technique of looking into the mind of the character, considering subtext & objectives and letting these factors control the acting, and the Larbin method of letting movement dictate the performance. Both are effective and interesting ways of going about acting, however I feel I identify more with Stanislavski's method acting approach, so I will try to apply it to my own characters, bearing in mind that the rules may not apply as the dialogue is mainly the thoughts, or subtext, of my main character.
Dr. Ben Sobel Caroline Action: Seated, deep in thought. Seated, talking to Dr. Sobel. Given: Daytime, psychologist's office, doctor and patient. Objective: Listen to patient's problems Seek help from Doctor. without letting on boredom. Beats: Change in tone of voice from Wondering if Dr. Sobel is awake calm to angry. or listening. Subtext: Hates his job, tired of listening Upset and distressed over how to people whine about their her current boyfriend is treating lives & loves. her. The subtext for Dr. Sobel's character is not simply guessed, as it is revealed in the sequel, 'Analyze That', that he only chose his career on account of his father. Research!
Later on in the week we had a guest lecture from a year 3 student on the motion capture software, 'IPi Soft'. This was a great experience as we could see first hand how motion capture is executed, as well as being given helpful tips on how to use the software itself, including calibration, jitter removal and using the 'Euler Filter'. Although I'd definitely want to experiment with the software and create some animation with it, 'Ipi Soft' is only really useful for capturing movement and action, and not so much for subtle performance animation. This is why I won't be using it for this unit's performance piece - my scene is driven by dialogue and my characters are seated. Below is a clip taken from our Mocap lecture, acted and filmed by my friend, Lloyd.
In place of a lecture we had one of our tutors, Amin, teach us how to make a basic mesh of a human head in Maya. Unfortunately I was absent for this session due to ill-health, but one of my classmates kindly showed me what I'd missed. I knew that seeing the process alone would not help me remember it for very long, so I decided to do it myself and document the process so I could make my own pdf tutorial for future reference.
In a follow up session the next week, we learned how to use the software, 'Sculptris'. Similar to Mudbox and Zbrush, 'Sculptris' is a modelling tool comparable to sculpting a sphere of clay, but digitally. We started by importing a basic head mesh into the programme and experimenting with the tools it has to offer, such as 'grab', 'smooth', 'pinch' and 'crease'. I used these to give my mesh some more characteristic features, such as high cheekbones and a cleft chin.
We were then asked to create five different facial expressions, retaining the volume of the head. This was great for studying facial expressions and learning how to manipulate specific parts of the face in order to create the look wanted. However, in a real working scenario it is more likely that facial expressions are achieved through rig controls, not the mesh itself. As well as this, the polygon count was incredibly high for something that would be used for animation purposes, not to mention the fact that 'Sculptris' favours Tris over Quads - both factors would result in many problems when imported into Maya. Therefore, although it was useful for practice, in future I will opt for Mudbox or Zbrush over 'Sculptris'.
We also had an ex-Rave student come in to talk to us about modelling, whose portfolio of work was outstanding. He ran a quick workshop at the end of his talk where he asked us all to model a washer. After a good 15 minutes of frustration and experimentation, he showed us his workflow, and we picked up some knowledge of which tools to use, such as 'bevel'. Although it's by no means perfect, I'm happy with the finished result, and the talk has motivated me to practice my modelling more.
