Now that I had one model moulded and hopefully ready to animate, I decided to do some tests with mouths that would match my birds. I started with the first bird I modelled, the red one, and used her colour scheme for the matching mouth sets.
Before I even started with making them however, I looked at all the possible ways Hayley and I could make our mouths. We have discussed this quite a bit, and we agreed that we were to make replaceable mouths, just because it would be much more easier for us to manipulate with our plasticine models, instead of having to shape and reshape the model each time for each frame. Also Hayley has experience of making replaceable mouths, and I really loved them, so we thought it would be best to do them this way.
I really loved Aardman's simple mouths, and noticed that other than those that are 3D printed, the replaceable mouths out there are very much the same, especially those made form plasticine. This meant that I had a lot of references and resources to look from, as well as of course using myself as a reference in terms of what my mouth looks like when I say certain letters.
For this test, I decided to try and animate a simple 'hello', purely because I was hoping this would give me a feel of this whole process, and how well I would be able to handle it. Also I didn't want to aim too high in the beginning, as usually when this happens if something goes wrong I am more inclined to panic and give up. Therefore I thought a simple hello would suffice for now.
In Photoshop, I made a quick gif of what I want each mouth position to look like for each sound. This was hard for me at first, but I felt that when I said the sounds aloud, I was able to use myself as a reference in order to help visualise each position better. This is how I got the mouth positions in the end, because trying to copy the existing ones actually really confused me as they didn't really look natural to me.
Once I had my own source to take influence from, I decided to get on with making those exact mouth shapes. This was an extremely finicky process because the mouths were so small. I used my model as a good reference for size, so maybe for the next models that I make, I will have to make them larger, as at times it was really tricky to do the little details, such as the teeth! However once I got going I ended up really enjoying this whole process. Though it did take me a long time to make them, I am pleased with how they turned out, and I think for my first time making them, they are not bad!
I then copied my gif from Photoshop, but used my mouths to make a mini stop motion test of 'hello'. I am so pleased with how this turned out! I think that with an actual voice narrating it, you would really be able to tell that it is saying 'hello'. The mouths on the left are just the mouths only, where as with the mouths on the right, I incorporated the beak. I think they both look good, so the only really way to see how they would actually turn out is on the model, so that is what I did next.
And here they are. I also like how these have turned out, and I am so pleased with myself, especially because I was dreading this whole process as I didn't think I would have the ability to make something like this. I did think however, that once I placed the mouths on the model, it seemed much more realistic, but much too fast. Perhaps this is because there isn't any sound to these gifs, but either way, I was worried that everything looked much too sped up. Also, these are just tests where I used my phone to capture the animation which is not what we will be doing for the actual animation, so hopefully they will look even more realistic slowed down, and without much jumping around! I will also have to be careful of moving the beak around too much also, as I noticed that it moves around a bit more than I would like.
Because the pace was really frustrating me, I quickly had a go at doubling some frames such as the 'L' mouth pose and the resting mouth pose. I think this definitely looks much slower and natural, but it will all depend on how the voices are recorded and how quickly the narrators are speaking.
In addition, I did all of these tests with the beak (left) and without the beak (right), just to see if the beak was really necessary. I think without the beak you can definitely see the teeth on the model a lot more, but other than that there isn't much difference. This is something that I think I will highlight in the crit, as well as asking Hayley for her opinion.
Then because I was really happy with how the mouths were coming along in conjunction with my model, I opted to go ahead and make all the other mouth poses to the sounds that I could think of. Some of them over lapped because I didn't know which pose suited which one, for example 'N' could be said with your mouth open or closed. But at least I am prepared if Hayley agrees to use them for our animation, and we will be one model towards animating!
I have surprisingly really enjoyed this process, especially because I was convinced I would be terrible at it. I think because I like to get things done efficiently, this took a really long time, so I was worried I would never get it done, but I took my time and I am so pleased with how the tests came out!
I then made some more mouths for my other model, using the same references and influences. I felt like the green is less appealing than the red, so perhaps I can consider this for the other models. It almost looks 'dirtier' than the other set of mouths, but thankfully this can be cleaned up with Levels and contrast in Photoshop.
I started with testing out the 'hello' on this character, and again liked the results. The one on the left includes an extra pose which I am not sure I like as much as the test on the right. I feel like it makes the model look like it has a lot of large, goofy teeth. This however might be nice for exaggeration and emphasis purposes. But I would be happy to stick to the test on the right. That one also would be great if our narrator was more Northern sounding, because of the 'o' sound looks much more broader matching the dialect. Again I made sure the mouths were in proportion to the models.
I also tried a new word with this mouths, which was 'goodbye'. I quite like this result, though I am slightly worried about those narrators who speak quickly. I think I will have to ask Hayley if she can get her narrators to talk a bit slower, as well as mine, because I think if they speak really quickly it is going to be hard to distinguish the words with the mouth poses, but it is something to bring up in the crit as well.
No comments:
Post a Comment