Joy Competition Animation

Only just got around to posting this now, but back in November, I entered the Joy competition held by Spline Bomb, in the end I didn't place, but it was a good exercise working without any sound. It was harder than I had anticipated as sound can really carry an animation and help it along. Voice or sound will have beats and you can normally animate to those, but without this there needs to be some sort of rhythm to the piece.

With the bear rig that I created, it worked okay, however there were issues with how much I could animate it without it being deformed. This comes down to the weight painting and if I had more time to spend on the rig, I would have liked to have made sure that it was capable of all the animation I needed. There are parts of the animation I'm happy with, but it needs some work. I think I would have found it easier if I could have constrained the hands to the bear, however I couldn't figure out how to do this without the arms deforming a considerable amount. I used FK on the arms instead, to ensure and constant contact on the bear, yet when it came to switching between FK and IK, I think this is where a few issues arose. I need to explore more into the switching and how to do it successfully and seamlessly, as right now in a couple of places the arm jumps quite sharply. I might try and revisit this at a later stage, and hopefully by that point I will have improved so I can go about fixing the errors.


Joy Competition from Siobhan Parkhouse on Vimeo.

Practical Element - Character

Through testing out motion capture and retargeting mocap data onto rigs, I have found out that I'm not able to get the amount of control that I would like from some of the rigs. Also with the rig that I have been using I still find that it is not plain enough for it not to have any contributing factors to deal with aesthetics. As I am focusing purely on the animation and performance I don't want the model or textures to affect this. To overcome this I've modelled and rigged my own character. I've tried to create a model that doesn't have appeal in the way of design or textures, and to make sure that it doesn't fall into the uncanny valley I've taken away the face and hands and replaced them for basic forms. Eye lines are one of the key factors to any performance, but it can also be the tipping point for something to be uncanny or not. The eyes need to be believable and not feel 'dead'. As there is no face it wouldn't look right aesthetically to have human like hands, by creating mitten hands it also allows me to spend more time on the mechanics of movement and the pure performance. The iPi software doesn't capture finger movements, only the wrist action so I would have had to add these all in myself using keyframe.

I used the HumanIK system within Maya to set up my skeleton. When you retarget the data to a rig, you first have to make sure that the rig is characterised and you do this within HumanIK, so it made sense to create a rig in this way. I've weighted all the joints, however I'm having a few issues with the controls. With HIK it will set up the controls for you, I need to go back in and check a few things out, however when I tested it earlier the controls weren't working as how they should be. I may find out that I will have to put the controls in by hand (which isn't a major task) the worst case scenario would be having to delete the rig and start again, but this time rigging it by hand. Hopefully it won't come to this and that I can get it sorted soon. Once it is sorted though, I plan to put my old mocap data onto this rig just to see how it works. I may still need to tweak it and adjust a few things at that point. I want to make sure the rig is working to the best that it can before I move onto my major piece of mocap animation.

Render Tests

I didn't think I would get to the stage where all my shots were ready to be rendered, however this is the case. The animation is some areas aren't as polished as I would have liked but the main 'storytelling' shots are the best I could of got it in the time given. To show my understanding of the whole process I needed to have at least some shots fully rendered out. I'm aiming to get it all rendered but with only 2 days left I'll have to see how far I get.

The image below is a test render I did for one of the shots (the cabin texture my friend created for me still needs to be added in, hence the black object). I changed the colour of the camera environment - this can affect the objects within the scene by adding a subtle colour tint onto them. Even though I'll be rendering each image out with an alpha, I thought I would put this on each camera as it gave me a good idea when I did test renders. I will be rendering out with mental ray as with the Morpheus rig, the skin textures are only visible using mental ray. This also allows me to have a higher polished shot, set on production quality. I have also turned on final gathering, keeping it at it's default setting in order to add in softer shadows and an all round good illumination to the scene. With the lighting, there is just one directional light in each scene with shadow rays turned on. Initially I had kept it on it's default settings, however I found the shadows to be too harsh so I increased the light angle and shadow rays to make them softer.

I really like how the test renders are turning out, the tree assets that my friend has created are great. They really fit well in the scene and the way he has textured them makes them seem soft and works well with the shadows and the style of the environment.

Animation Process

The main focus for myself in this module is the animation, I wanted to push myself further and work with more than one character on the performance and acting of the shots. As I've been working through the shots I've found it quite a challenge in some respects, as I've never really pushed my poses before and trying to achieve more performance is not as easy as I thought it was going to be. I took reference videos of myself but that in itself is something I need to work on. I need to exaggerate more when I do them, I think it's more of a confidence issue and being able to free up and get into the role of my character. In that respect, it's more like acting so I need to be more accustomed to the whole process.

Working my my pre-viz I had worked out the rough timings of each shot and from here I blocked out the main poses in stepped mode, before changing to linear to add in more of the subtle details and then onto spline. Some of the shots that I have been working on have taken me a lot longer than anticipated and has set me back quite a bit on my schedule. At the minute I'm just working through each shot (the majority are all blocked out now) and polishing up the animation. As I've been working through them I've realised I could cut a few of the shots out which will free up some of my time, but also they weren't really needed as they tended to be walking shots. Walking isn't storytelling unless an emotion is portrayed through it, a lot of the time you don't see characters walking to their destination. Instead you see them about to embark and their destination in sight, then in the next shot they arrive. If I had more time I would have liked to keep the shots in, then make the decision when it comes to editing, but at the moment, I have more important shots to complete which attribute more to the storyline.

The animation below is a test I've done with sound. It took me awhile to get the slap sorted, I took a reference video, but also looked at other examples. I think the sound adds more of a believability to it, as without it seems flat. I've added in some extra details, such as the movement/fidgeting when the man is sleeping, I found that without this it was lacking and there needed to be something else. I hadn't really thought about putting this in until the other day and for me, it really makes a difference. I think it still could be improved but for now, I think I'm going to move onto another shot. I find that if I spend too long on one shot I end up adding things that don't need to be there, it's best to get a fresh eye on it on another day. I want to keep adding these extra details in within the other shots, but only when necessary. Before this module I've never really gone into this detail, I've just concentrated on the main pose and left the rest of the body. It's something I need to practice more and get my head around as I always find it quite overwhelming as there is just so much to think about and do, what with overlapping and secondary motion, the little nuances to make it that much more believable. It's a process and hopefully it's something I can keep improving on and will get to a stage where I feel comfortable and happy with what I've produced. 

 

Animation Practice

So I thought I would get in some more animation practice as I had some spare time just before and during Easter break. I wanted to get a pushing animation done as this is one of the basics and something that needs to be practised. I made sure I created reference videos so I could analyse these and refer back to them when animating. Reference videos is something I always do now, they are so useful, trying to do it from your head, I find just doesn't work. There may be parts of the body that move without you really noticing unless you analyse that video.

From this I broke it down to key poses which I started creating within Maya. I decided to use the Malcom rig, courtesy of AnimSchool. It's a really good, comprehensive rig that I've slowly got accustomed to. There is a lot of flexibility within the rig, stretchy limbs, IK/FK switch, good amount of facial control and the ability to use abxPicker (unfortunately I had issues with trying to get this to work within my Maya so couldn't get to try it out).

I really took my time with this animation to try and get it as good as I could, whilst I had the time. I made sure I got feedback from several people whilst in the process, to see what I could improve on and if the timings were correct. As you can see below, that's my blocking stage before I went into linear. I followed the same workflow as I did in the visual effects module and can really see the benefits of it. Below the blocking is my final animation, I spent the time cleaning up the keys in the graph editor and making sure the curves were smoothed out. This extra time does help, it smooths out the animation and makes it more fluid. I lost some work near the end and had to redo a lot of it, but it was for the best really as I ended up doing it better than the last time and in a quicker time. I've learnt a lot by doing this, and got to use the IK/FK switch for the first time. By changing the arms to FK, it allowed me to have the hands stay in one position whilst the rest of the body moved. The next thing I want to improve on is rendering, I want to render my animations out nicer, for example how Animation Mentor do it. By the looks of things it might be ambient occlusion or final gathering, but I will look into this further soon.

Render Layers

A couple of weeks ago we were introduced to render layers and how they would be used. Render layers allow for more control in post production, as there are normally many different pieces brought together in vfx it's good to have this control as things may not match up perfectly with one another and will require adjustments. They work in a similar way to how the layers do in maya, by putting different objects on different layers it allows you to render out those specific items. When it comes to render layers, you typically would render out a colour pass, shadow pass, reflection pass and an ambient occlusion pass. On each new render layer you create you can apply overrides onto them which will just affect that specific layer. One of the main uses for overrides is the naming system, another one would be for an object to receive shadows but not emit any.

For my scenes I will only be doing an ambient occlusion pass as this will help the rendered shots sit better in the scene. Ambient occlusion is self shadowing and just gives it that extra bit of depth and realism. I thought of maybe doing the other passes but I didn't think it would be that necessary, any adjustments to colour I could easily do within After Effects itself, the reflections are all set up to how it should be within Maya anyway and as for shadows there aren't that many due to the floor being black. For a scene that was more realistic I can see how these passes would be really beneficial, it all adds to that realism and helps the overall composition. As you can see below this is one of my frames as an ambient occlusion, it gets rendered out with an alpha channel so that the background will be transparent (this is just a screenshot so the transparency is replaced with grey). You can see the shadows on certain parts of the alien and the tree and once put together with the final comp, it really helps the alien shot sit better within the scene.

Maya Set Extenstions

Because my alien character was created in Maya I also created an environment for it to exist in, this would be my virtual set extension where I would place my green screen footage of Jeff into it in post. As you can see the environment itself is very simple, however once rendered out it looks a lot better and if there was too many objects within the scene it would have felt too cluttered as I wanted to create a clean cut look. As I have mentioned in a previous post, the background would just be a texture dropped behind the footage in After Effects. The image below shows a plane for the background but this is just for the lights to bounce off of and produce reflections.

When going into post with a virtual set and character you have to make sure both the models and the real footage work together as a whole. They both have to look like they have come from the same thing for it to be believable. In order to achieve this as best as possible some colour grading needs to be applied, tint is a good one to use as you take the blacks and whites from the scene and can adjust your footage to it accordingly. I did this to all my shots and I also turned down the lightness of my alien character as I found that it was a bit too bright when put in the shot with Jeff.

I also had to do some other adjustments, especially with the ambient occlusion layer as this tends to leave a harsh outline around the objects. To get rid of this and to create more of a seamless integration, I used a simple matte choker to take away that harsh edge by a few pixels. By doing this it already makes the character sit far better within the scene. With virtual set extensions you also have to make sure the lighting matches how you set up the lights when filming green screen. Ideally you would set up the lights when filming to match your virtual scene, by doing this it makes life a lot easier. Trying to put in a light in post would be a lot harder than doing it all correctly in the beginning.

Animating the end shot

For my last shot my plan was to have my alien character jump onto Jeff, our actress. When it came to creating this animation I realised I would have to find a way of having that section of film footage where Jeff falls over and putting that into Maya. I needed to do this as I couldn't just animate it without having the footage as reference, if I didn't have it the animation wouldn't line up when it came to compositing the two elements together. In order to do this I had to take my video footage and export this out as an image sequence. It's much easier to work with image sequences in Maya than it is video. With this image sequence I then attached this onto an image plane as a shader, I could then adjust the image number attribute (as seen on the left) to delay the start of the image sequence so that my animation would work with it. By doing this I could make sure everything lines up with one another and the timings of both the animation and video are synced with one another. All I would then have to do is line up both together within After Effects.

The screenshot below shows the setup I used within Maya. I made sure that I had my rendercam setup and locked off on the right and my perspective / workcam on the right. This allowed me to move around in the world space but also see the effects within the camera view. Because of the way Jeff falls it was hard to animate the movement of my alien, it couldn't achieve and nice full arc. This wasn't Jeff's fault as we couldn't really have someone running and pushing her over at force because she might hurt herself and the crash mat we had wasn't anything substantial, instead she fell over in a way that she felt comfortable with and even so it's a good enough performance. The arc may not be there but I made sure the rest of the animation was up to scratch.

It was suggested that maybe I move the alien back, towards the right to give it more space to jump, however the previous shot shows the alien jumping up close to the camera so by having the alien further away from Jeff it wouldn't make much sense and there would be a delay in the time it takes for the impact. It's not perfect but I still think it fine, you have to adapt and work with what you are given with and I feel like I've done my best with it. In post I will go in and mask out one of Jeff's legs in order to bring it infront of the alien and also add in a shadow for when the alien jumps onto her. This will make the character sit better into the scene, rather than just looking like a 2D image.

Animating a Quadruped

I realised I was being quite ambitious, wanting to animate a quadruped when I have only worked with a biped a couple of times. My rig that I created for my alien character wasn't as accurate as I would have hoped, I ended up rigging the front legs wrong so that it bent the wrong way. When it came to animating I just had to work around this and do the best I could with what I had. I looked at a lot of reference videos to get a better understanding of how quadrupeds walked, especially cats in particular. I then took one video and broke it down into the main poses, contact, down, passing, up, contact. By doing this I had a clear image of each pose and could then work from these to block out my animation.

As my alien was jumping down from a tree I found a really useful video that I ended up referencing a lot. It was really helpful that it was in slow motion as well as I could take my time going through the video and analyse each shot and see how the body reacts through the air and on impact. There's a lot more to take into consideration when animating a quadruped, the legs have to work with each other yet they are offset at different points.

I also decided to follow a proper workflow this time, in the previous module I had tried to block out my animation first and then change this to spline, however I had a lot of issues with it and ended up just working straight from auto-tangent. This time I wanted to try again and I actually succeeded, by blocking out just the key poses first and not worrying about timings it makes things a lot easier. You are able to just concentrate on building good poses and putting in key ones to build your foundation. From here I then went in between and put breakdowns, the computer needs as much information as possible, without the necessary information it will just guess. Which is the issue I had before, I wasn't detaield enough with my blocking. The more time you spend in blocking, the easier it is when it comes to clean up. Below is the final stage of the tree jump animation, I'm really happy with this as I feel like I have a better understanding of the whole process and can progress this further next time. Blocking, adding in breakdowns and then going into spline is a far easier and cleaner workflow than what I was previously doing.

Rigging / Blend Shapes

As my character is a quadruped, I knew the rigging set up would be slightly different so in order to get to grips with this and create a fully working rig, I followed tutorials online which gave me a basis to work from. My previous knowledge of rigging still came into use with mirroring joints and putting in IK spline. I found that researching into a cat's skeleton was really useful as this showed me where to put the joints and also the names. With cats, they have an extra bone on their front legs above what essentially would be the ankle, this allows more movement and flexibility. Because of this set up, the placement of the IK's would be slightly different to how a biped would be, so I added in an extra IK between the joints to allow for this movement. I realised that after I rigged and weight painted the model, I could have done with less joints in the neck as a few of these aren't really needed and it would have been easier to weight paint with a couple less. I originally planned to include joints for the eyebrows as this would allow me more control over this area, however I ended up forgetting to put these in and I would have had to unbind the mesh and do it all over again. Instead I had created blend shapes for the eyebrows, including the main expressions I will need.

Using the component editor was something I came across within the tutorials and found it a really useful tool. By going into component mode you are then able to select certain verts and adjust their influence. This is a more accurate way than just weight painting and allows more control, obviously this should only be used for small areas, areas where it's hard to weight paint or where some adjustments are needed.

When it came to creating the blend shapes I came across a major issue, when using the blend shapes they ended up distorting the mesh in areas that were not modified in any way at all. I was told by my tutor that some way along the line the mesh got distorted from the original one, and seeing as I duplicated the mesh before I bound it every time the blend shape is used, it always wants to go back to how it was, hence the deformations in other areas of the geometry. To try and fix this problem we first unbound the mesh from the rig and exported this out separately, by doing this I then had the original mesh back. I then imported it into the scene I was working on and began to detach all the heads of my blend shapes from the body. Seeing as only the tail and right front foot was affected, I would not have to re-do all the blend shapes again. After this I then re-attached the heads to the new body. At first this seemed like it would work, however in the end the blend shapes became destructive. We found that only a couple were really destructive, so by taking those away and duplicating a blend shape that worked correctly, we used that as the new geometry. This process took longer than I anticipated especially seeing as I had to re-weight paint my character and attach all the controllers again, however it is still better than having a broken character! I have learnt many new things so it wasn't a waste of time, and also by having to re-do certain aspects I understand the process more and am able to do it in a much faster time than before.

Modelling my alien character

I started modelling my alien character about a week ago and I have now gotten to the stage where I am just adjusting the topology to ensure the lines flow correctly, this not only produces a cleaner model but it will also help the rigging, weight painting and animation stage. Modelling a quadruped was something new to me so I made sure I looked at reference a lot of the time to get a better understanding of the shape and build of an animal. I really benefited with the advice I got from my friend Jacob as I would send him images of my model and he would give me feedback of what I could improve. I also learnt a lot of new tools and techniques with the modelling workflow e.g. duplicating half of the model as an instance, this allows you to adjust one side of the model and it will do the same on the duplicated half. This is a good way to work as you can constantly see the transitions you make and don't need to keep mirroring and deleting faces. The sculpt tool in Maya is a really quick and useful tool which helps you even out the edges of your model, it makes it a lot cleaner, before I did this the mesh was quite messy and I would not have been able to take it much further without cleaning it up first.

You can see the stages of development from the pictures below and I'm quite happy with how it's turned out, it still needs tweaking here and there before I am able to move onto rigging. As you can see the tail was originally down but I was advised that for rigging purposes it would be easier if it was modelled straight. The one thing I could have done different is the way I attached the legs, ideally they would have been attached to the side of the body and the edge loops would have flown a lot better; but instead I connected them to the bottom of the body like I would with a biped character. Hopefully this shouldn't impact on it too much but I shall have to bare this in mind for next time.

Compositing Test

To make my first shot more interesting I decided to add more than one spaceship into the scene. I began with masking out the foreground and background sections in order for me to be able to place the maya sequences in between the two. Our original footage, the sky was quite grey and bland so I thought it would be nicer if there was a blue sky, again this was just done by masking out the certain area and I used a sky texture from cgtextures.com. To begin with the sky was too vibrant so created a ramp and an adjustment layer to counteract this. I also had to do some colour correction on the actual shot itself for it to be more of a seamless integration.

As I will be using a shot as a image plane within Maya, (to line up the animation so I know what's going where and how fast it will be travelling across the screen) I placed just two coloured solids into the shot so I would get the basic layout of where I wanted the spaceships to be / their ending position. From here I rendered out one frame and attached it to a new camera in Maya. Using this image as reference I then animated a few spheres adding fur to one of them to get an idea of what it would look like. The lighting still needs some work, at the minute there is 3 point lighting on it but the intensities may need some adjusting, especially around the edges of the fur ball. As I have mentioned in a previous post, I had issues with making the fur dynamic, in the way that the fur would always pierce through the geometry even though a collider was set to it. Apparently this can't really be solved the only way around it is for the object to be moving really slowly which would reduce the risk of the fur going through the geo. It was suggested to me that a way I could possibly get around this is by setting out the animation how I would like it within Maya, then applying the fur, slowing it down to where the fur doesn't puncture the geometry and then try speeding it up within After Effects. In theory this should work but I would need to test it out first, speeding up clips can sometimes create interlacing in the footage, depending on how much you speed it up by.

Blend Shape Tutorial

I had previously tried out using blend shapes during the summer break so I had a basic understanding of the process. With blend shapes you have to duplicate the master geometry and alter the duplicates by adjusting the vertices. There are many uses for blend shapes, the main one being a deformation, the other use is corrective blend shapes. This is useful to correct certain areas of the geometry, for example when you open your mouth, there are deformations that take place in the cheeks, so by using blend shapes you can put this in place to add more realism. They can also be used to alter any areas that aren't working correctly. You have to make sure that you don't delete the history or freeze the transformations of the duplicates of the geometry, if you do you the blend shapes will revert back to the original place of the master geometry. When selecting the blend shapes, always select them in the order you created them and then the master geometry last. From here you would create the blend shapes and add set driven keys to allow you to animate it.

I will be definitely be using blend shapes when it comes to my character, most likely in the form of expressions, I will have to make a note of all the expressions I want and duplicate the necessary amount of geometry needed.

Fur Test

As I want my alien to have fur I thought it would be best to try it out within Maya first before I went too far. By doing this I will get a better understanding of what I will need to do when it comes to the final model. The images below are just using the presets given, there are a range of different types all of which you can combine with one another by blending or having the preset just on it's own. I have just used a polygon sphere to test it out, but when it comes to my alien I will have to be careful when it comes to UV mapping as this can affect the way the fur is placed. By selecting the specific geometry all you need to do is add a new fur descriptor. Within each fur descriptor there are many attributes which you can edit, all of which will change the way the fur is, for example length, density, colour, baldness, roll, curl and others. Again for the sake of testing it out, I have just used a preset fur description and only altered the length of the fur when it came to experimenting with dynamics.

Below are some of the other presets available within Maya, the last few I have either adjusted the length or blended a couple of preset furs together. At the moment the preset, Polar Bear is looking quite promising and I think I will develop on this further by adjusting certain attributes to see what effects I can achieve. I have also began to think about my spaceship and after seeing what fur looks like rendered, I'm liking the idea of the ship to just be a big fur ball. I have been having trouble trying to come up with an idea as there was nothing that really correlated between my alien and just a normal spaceship. There needed to be some resemblance and I think fur will be the best bet.

The videos below are tests of animating the fur dynamically. I followed a tutorial which I found online (HERE) and decided to give it a go to see what could be achieved. The first video is with a longer hair length, set at 2 whereas the one below is set at one. By following the tutorial I had to adjust a couple of attributes and created a hair system to my polygon sphere, by making collide the hair system and the sphere it should stop the hairs from going through the sphere, but as you can see in the tests this doesn't seem to be the case. I will need to look into this further and possibly try out a different technique, but for now I just wanted to test it all out and from here I can develop it further.

Animating

I did all the animation within Maya and in a previous post I mentioned that it took me a long time and a lot of different iterations before I was happy with a run cycle I had created. Throughout the whole animation process I have constantly been going back through my reference videos we took at the start to work out how our characters were going to move. I also would act it out to myself to understand the weight and movement of the character more.

To begin with I blocked out the animation which gave me a good idea of how long the shots would be and I could work with that when refining the animation. I was then going to change the curves from stepped to linear and then to spline, however I found this really awkward, when changing between them my movements ended up being all over the place and I couldn't edit them too well. I'm not sure if I keyframed everything each time so this could have been the reason why so I may try this technique again at another stage when I have more time. Because of this however I ended up starting from scratch and using the auto tangents. I didn't really edit the curves too much once I keyframed it all, maybe a few tweaks here and there for timing but I didn't really have the time to go in and really analyse each part. I would have liked to as this would have given me more practice within the graph editor. I did keep referring back to the blocking animation though to make sure I was on track with the timings.

The video below shows a playblast of where Gloria falls into the buildings which consequently knocks them all down, because I couldn't get any reference for this I had to guess and luckily it turned out really well on the first attempt. I made sure they fell one after another, but at a quick enough pace to look convincing. I think they all fall over a couple of frames which may seem really quick but it needs to be that fast, needs to be punchy. I love animation and even though I struggled with the run cycle it just takes time and practice to get it looking right, by doing this module it just confirms that I do want to be an animator and more specifically a character animator.

Run Iterations

I spent a long time working on the run cycle, probably over a week and it got to the point where each time I edited it, it would just get worse so in the end I think I deleted it about 4 or 5 times to start afresh. I took reference videos of myself running in similar style shoes to our character, I even looked at references off the internet and other students work to get a better understanding of how they were working. I even got feedback from another animator yet it still wasn't right. I just wasn't happy with the way it was going. In the end I finally got out 'The Animator's Survival Kit' and as soon as I did this it all just clicked into place. I took screenshots of the main poses of one reference video and used that to create my key poses of the animation, however I think because our character was so stylised and her proportions aren't of a normal human being, the run would have had to be adapted in some way. That's what is good about 'The Animator's Survial Kit' it shows you many different iterations of runs and walks, some stylised, some more humanised; it even highlights certain aspects of the cycle that I would have never even thought about, such as offsetting the timing so everything isn't moving at the same time as each other. By doing this one little thing and offsetting by just a couple of frames it changed the look drastically.

Walk and run cycles are always going to be the hardest things to animate as everyone knows how someone walks, so the smallest of things will stand out to the normal eye. The only way I'm going to get better at it is just through more practice, and looking at more reference and really sitting down and studying it before I go ahead and animate. Throughout the whole time I was animating I kept getting feedback from others but the problem was that I just got told the same things I already knew and it was hard to see how I would go about to change those factors. Either way though feedback is always useful in whatever you're doing as it can point out things that you may have never of noticed beforehand.




This last video is where I pretty much gave up and scrapped the whole lot to start again. Sometimes it is needed though, just reworking the same thing over and over again was only making it worse, starting again allowed me to have like a clean canvas where I could just concentrate on the movements from scratch and not have other things distracting me. I think next time I will definitely draw out thumbnails of the main poses so I know what I'm working to and look into doing dope sheets. Dope sheets help with organisation and timing of the animation and are really useful, I look back on it now and I realise how much use it would have been to me if I had actually done this.

Blocking Animation

In industry, animation is normally blocked out first to put down the key poses and to get a sense of timing. If anything in the script was to change, this would be the ideal time to do it. II thought I would try out this method as before I was just keying the animation using the auto curve and adjusting it from there. Everyone has their own way of working but I thought I may as well try this technique out. Blocking gets across the idea and motion of the character easily, I have showed the rest of my group the below video to get feedback and before I change the curve from stepped to linear I will adjust a few things, such as slowing down the run cycle. I also need to add in some facial expressions, blinks and just general small things that will add to our characters personality.

When I go to change the curves to linear, I will be adding in extra key frames to smooth out the animation and maybe deleting any unnecessary ones. After it has been completed with linear curves, I will then take them to spline curves, this will be the last stage and hopefully the refinement of the animation where I will be adjusting things here and there.

The below video is the blocking animation, not all of it is there as our second character is just being finished off at the moment. To utilise our time the most efficiently, I have begun animating with a non textured model and environment. These are not needed for animation and it means everything is progressing and nothing is on hold. They eye texture is present however as I wasn't able to gauge the eye movement very well without having anything on the eye geometry.


A useful blog I came across awhile ago was, 'Animation Tips and Tricks' where professional animators give their insider tips. It's really helpful and gives you a good insight to the industry practice. One that I found was on blocking animation and why they do it within the industry, Animation Tips and Tricks. It gives the supervisors and directors a clear idea of what you are going to be doing and what is going on in the scene. One thing I haven't done, but will do in future is before I go in to blocking the animation I should really draw out thumbnails of each shot to give my ideas more grounding and stability. Looking at this blog and back at my blocking I realised in some places I could have not included as many key poses as it was clear what the intention was. This was my first go at it so there are still lots of things I can improve on, but at the end of the day it's a learning process.

UV Mapping

I had previously unwrapped Gloria - our main character, which took me a really long time, probably longer than it should of, but I got there in the end. I knew in some places it was a bit messy and the way I unwrapped parts probably wasn't the most efficient way to do it. I had to re do some of the maps however as when it came to smoothing our character she turned out to be a fair few too many polys. Ideally a character should be as low as possible around 6000 - 7000 polys, trying to be the most economical. Because she was so many polys I went and started deleting unnecessary edge loops and by doing this I got her below 8000, the issue was that by deleting edge loops I had messed up the uv map. I think by deleting edges after it was mapped I may have deleted the seams which would distort the map itself. I was annoyed at the fact I would have to re do most of he unwrapping again, but in hindsight it was a good thing. This time I was able to make the map a lot neater and more understandable.

Now knowing this, with our second character we have smoothed him and then Lija deleted the unnecessary edge loops first before unwrapping him. This shouldn't distort the map in any way as we are doing it after the editing stage. We just hope that smoothing it beforehand won't be detrimental in any way as we were told to unwrap first then smooth. I guess we shall soon se if there are any issues, but hopefully they would be an easy fix if anything.

Modelling our second character

Carrying on from the previous blog post, I have now completed our second character. He has turned out really well and I particularly like his facial features, he is such a stark contrast to our main character and I think the contrast will work great when they are put together. Due to time constraints I am not going to be able to UV map and rigging the second character as I know the animation is going to take me awhile, so I have passed these jobs onto Lija. Learning from our previous character we are going to smooth him first, delete unnecessary edges and then UV map him. This character has been quite simple to model compared to Gloria, however the hands were a bit of an issue, as he is quite 'chunky' and stubby I had to work out a good proportion for his hands, hopefully they'll be alright when it comes to rigging and weight painting.
 

Alien Animation

As I mentioned in a previous post, the next stage to my alien would be to animate it doing a walk, jump and an idle which would then be taken into Unity 3d for a turntable. I have created a playblast of the animations in each state of the alien; skeleton, skeleton and controls, geometry and textured. This gives a better overview of the alien and the construction that took place to create the final version.

I still find walk cycles a bit awkward to do and it is something I really want to get better at so when it comes to animating the characters for our cinematic; I will be wanting to create some reference videos to work from and analyse. I find it easier to do other actions and a bit of facial animation so hopefully this will stand out in our cinematic, but as most of it is walking I need to make sure that there are no issues with it and that I am able to execute it to a high standard.


Alien Animation from Siobhan Parkhouse on Vimeo.