3D Animation Project

Yesterday I have finished rendering the animation, added a background sound and my name and contribution to the video in Sony Vegas (as seen in any 3D Demo reel available online) and submitted it, one day ahead. Rendering the animation took way more than I anticipated, especially since I had to render it a second time because of the unsatisfactory results on the first render. Although the overall video quality of the demo reel is not that high, it's way better than the first one where the noise levels were unbearable due to the higher threshold. Rendering the animation for a second time took about 19 hours on my personal computer and 15 hours the first time I've tried. Had I decided to go for a higher quality, it would've easily taken over 30 hours of rendering. I believe the overall quality is not an issue for now, particularly because the project could be rendered again on a better performing machine, at a higher quality. That being said, when I started the dissertation...

The run is the final segment of the animation demo and also the one I considered the hardest to work on. Although I have put time and effort into making this running animation look at least decent the results are somewhat underwhelming and it's now too late to sink my time into it anymore since the submission deadline is getting close (I do plan on submitting the artefact tomorrow, one day ahead). I will however discuss the iterative process I went through and the issues I have encountered. The most relevant issues I identified were, again, the wobbliness of the character when running and the timing. Because the timing was the most crucial and because it wouldn't have made sense to re-adjust the poses without getting the timing right first, I've started from there. One of the principles of animation, as I have previously said, states that humans for example don't reach full speed in an instant when running, nor do they come to a full stop instantly either. However...

The confident, bold walk is the third and last walk segment before the running animation. Here, the character adopts the personality of a celebrity or at least someone from the upper class in a society. The steps are long, knees are swivelled and feet are twisted towards the exterior. The chest is all the way up front, the shoulders are raised and the back is straight with the arms bent and fists clutching, executing an arching motion cycle.   The walk strongly suggests superiority, confidence and boldness. That is the personality I wanted to convey through this walk animation.   As I have previously said, this animation segment was the best realised one of them all and only required minor changes. The only issue I was able to identify was the simplicity of the walk, the lack of any secondary action. The character’s head was not moving throughout the entire animation segment, fact that made the model look rather stiff or rigid. I have decided to add to the personality of...

In one of my previous titles ‘ Animation progress status and observed issues ’ I have identified and discussed about the apparent issues found in each of the animation sets. In the last blog post I have addressed the problems found in the leisure walk animation - so naturally I will now discuss about the changes I’ve made to the sneak walk animation as part of the iterative (refinement) process. But first, I’ll briefly list the identified issues: ·          Excessive movement of the pelvis during certain frames ·          Excessive hands movement ·          Lack of any secondary action Here is an old render of the sneak walk before making any changes to it: Starting with the pelvis movement, as anybody could have certainly noticed during the 5 th and 6 th steps the pelvis (hips area) of the model rotates up and down in an unnatural fashio...

In the next few blog posts I will be discussing about the problems I have identified in one of my older blog posts (Animation progress status and observed issues) and how I have solved them over the past few weeks. The posts will each be focused on one of the animations separately, starting with the leisure walk animation in this post followed by the sneak, the bold walk and ending with the running animation. I believe that by organising the blog posts in this way will make it easier for me and anybody else to refer back to if necessary, as well as keeping the blog posting constant, tidy and succinct. The leisure walk The issues I have identified last time in this first part of the animation chain were that the walk is rather wobbly and some of the steps are unequal. The seemly unstableness of the model when walking was caused by the excessive movement of specific parts of the body. For example, upon taking a closer look at the first 20 or 30 frames I have noticed that ...

Since last week’s post I have been making progress on the animation and prepared the scene for rendering. I will however focus on the scene preparation in this text and on the animation part in a future blog post. My decision to take a short break from refining the animation in order to prepare and organise the scene for the final render was based on my personal preference and working habits. Therefore, the reason was that it allowed me to better estimate the time I have left to complete my project (or any other task) before the due date. Now, there are exactly two weeks left during which I can further refine my animation because I have already prepared the scene for the final rendering. This not only gave me a great sense of progression, but it also allowed me to better plan my next moves when it comes to the animation. I was also able to estimate the time it will take to render the whole scene that now features ray-traced shadows, a three-point lighting system and panels. As a m...

While I was animating the character I had some renders done so I can watch my progress and observe any mistakes I have done or on-going issues. In order to do so, I needed to set up a camera so I can track the whole animation of the character. In this post I will present the process of setting up the orbiting camera I have used in my renderings. Initially I have looked at some online animation demo reels and I noticed that most of the cameras are either completely static either move with the target. The static ones capture the whole action from a side-view most of the times and the mobile ones can follow the action on any axis, moving accordingly. As the walking distance covered by the character in my animation is rather large and it would be hard to fit everything in a static camera’s field of view I decided to go with a mobile camera that follows the character. Prior to that, I remember looking into the camera view modes available in 3DS Max and seeing the pan and orbit o...

For the past weeks I have been working on the blocking stage of the animation process. In the last blog post I had about 128 unfilled frames (most important positions, no transitional frames) and now the blocking stage is almost done at 407 frames. Although far from being complete or polished enough, the animation is now starting to shape. In the last post I was also talking about how I have changed my approach on the deliverables of this project in terms of animation composition. As I have started the demo with a walking animation and then had the character transition into a form of suspicious, sneaky walk, I have decided to have the model perform a few walk types that all display a different persona. The first is a natural walk, where the character peacefully walks along. The second one is when the character walks as if sneaking into a building or behind someone’s back, treading carefully and being extremely cautious of the surroundings to not alert anybody. In the third on...

Since the last blog post I have started the animation process after briefly checking the skin deformations one more time and conducting basic scene management preparations (organising layers, hiding unnecessary bones and goals, etc.) before finally getting started on the animation. As I have mentioned before, the aim of the project was to create a walking animation, a pose animation and a jump animation for the chosen (human, biped) model. The animations would’ve been presented separately by having the character walk first, then strike a pose and ultimately perform a jump of sorts. The issue was that no matter how hard I looked through other artists’ animation reels, I couldn’t find such simple and rather ‘secluded’ animation cycles. Most of the reels told a story, or multiple ones. The idea is the character was never ‘just’ walking or jumping, or striking a 3 seconds pose and then that was it. Most of the reels had the character/s perform multiple actions, they had the charact...

This week I have successfully rigged the hands of the model and then prepared the rig for animation as well as cleaned the scene.             Rigging the hands was a straight forward task that did not involve anything new that I haven’t done before. I have started by creating a controller for each of the five fingers and aligned it with the third joint (the closest to the hand) on each finger. I have then linked the third joint with its newly added controller on each finger. The controllers were then linked to the hand bone so that the fingers remain attached to the hand (and move with it), but also independently.               Similar to the other controllers, I froze the transform and then locked their movement and scaling on the XYZ axis. The fingers should be able to go up and down (Z axis), left and right (Y axis), but they shouldn’t be allowed to rotate. Locking the rotatio...

Last week in part 2 of the rigging process I have taken care of the spine of the model by adding custom controllers and constraints. This week I will be rigging the arms and clavicles. I have started with the FK rigging first (as the rig created features both IK and FK arms bones) on the left arm. As in the previous posts, I first created three different geometric shapes and aligned them with the joints: shoulder, elbow and wrist. I have then linked and then constrained the orientation of each FK bone to that of its controller: ·          Wrist bone orientation to the wrist controller ·          Elbow bone orientation constrained to elbows’ controller ·          Shoulders’ orientation to its controllers orientation Now I was able to rotate the FK arm bones using the individual controllers. I have also locked the movement of the controllers on the...

In a similar fashion to the work I’ve covered last week, this week I have taken care of adding custom attributes and controllers to the upper body. At the end of this work session I had the spine, neck and head of the model rigged. The control shapes were the first to be added. I needed five different controllers affecting five different bones, some independently, some affecting two or more bones at a time. As seen above, these controllers were in order: COM controller (or the Master controller) – This controller serves as the main one: all the other (body) controllers have been linked to it. When moved, the whole body follows. The same is valid for rotation. Hips controller – This controller takes care of the hips movement (or the spine base). When rotated, the pelvis follows the rotation along with the hips. However, the first spine bone should rotate to a degree as well as a result. To get that effect, ‘Parameter Wiring’ has been used: a new Euler XYZ rota...

Having skinned the character, it was now time to take care of the IK and FK chains blending and add the (custom) controls for the rig. Adding a set of manipulators (controllers) is important because it ‘…makes it easy to animate by the animators’ whilst not adding any would result in ‘…making the pipeline of character animation more time-consuming, difficult and problematic’ (Bhati, Karbasi, Mahesar and Waqas, 2015). It all comes down to how easy an animator can control individual bones and bone chains, as instead without any controllers even selecting a bone would take much longer. However, the subject goes lengths beyond that with certain bones having bound to them certain reactions, constraints (limits) and more. I have started the process by freezing and hiding the mesh first, as using or interacting with it was not necessary for this task. Following an extensive online video guide I’ve found, I have started by creating an IK chain (using the ‘HI Solver’ option under the ‘Ani...