Works in Progress

#41

Thanks dinocarl! Sweet, glad you liked the battle pod… I was sittin’ around looking for something to build when I came across an old Robotech book I had on the shelf. I was about to build that transformable Veritech Fighter but got side-tracked with work. BTW, that VF you have built looks good, I see it transforms too, nice touch. Later on I may try building Raidar X or The Spartan and include it in an up-coming demo I’m trying to produce. If you animate a war sequence with the battle pod, be sure and send us a link.

#42

Ok. Here is a head that I am working on. Be harsh.

Oh, and does anyone have any tips for making the inside of a mouth? Last time I felt like a drunken dentist with no training.

#43

Originally posted by alfiebabes
[B]this was originally posted on the lip-sync mini challenge as

http://members.lycos.co.uk/aphil/movies/beef.mov
sorenson3 Quicktime under 2mb

the newest version (half re-done) is

http://members.lycos.co.uk/aphil/movies/beef02.mov
sorenson3 Quicktime under 1mb

comments and crits appreciated, I can’t quite workout the “lag” I seem to see in the 2nd version - it’s only an 850k file but plays real slow and jumpy on my PC at work. Even with 512mb of RAM!! [/B]

Not sure what the problem is w/ the lag. It seems to lag a little on this machine, too. What I do when I finish setting keys for the mouth is to take all the keys and slide them back (earlier) one or two frames. For some reason our brains tend to pick up on the visual slightly before the sound, so putting the visible shape slightly before the sound that accompanies it will look more in-sync than if the shape exactly matches the sound.

Re: the animation, it looks a little better, but still has a lot of the “poppy” problems I mentioned in the other thread. There also isn’t much shape blending, which is something else mentioned previously. Try working on those two issues in your next pass.

Also get a mirror and watch yourself speaking this line. First watch your jaw for a while. The motion of the jaw isn’t always tied to a particular shape of the lips. Also, the jaw is often where the accents in a line of speech are most noticeable, opening wider for stronger spots and much less for softer places.

While this is harder to do, notice the sounds that are made in conjunction with your tongue…things like S, T, N, G, K, etc. Even if you don’t have a tongue in your character, it helps to notice that for sounds like these, the tongue is pressed on the roof of the mouth to some degree, which is difficult to do if the mouth is open very far. Basically, don’t have the jaw open very far when creating shapes for these kinds of sounds.

Something else that I noticed w/ your model while looking through the clip frame-by-frame is that you have some spots where the lower teeth are moving up and down, but the jaw doesn’t move. This shouldn’t be happening because in real life, the teeth are connected quite solidly to the lower jaw. When the jaw moves, the teeth go along for the ride. Addressing this issue should help you create more effective poses.

Keep working at it! :slight_smile:

Justin

#44

http://members.lycos.co.uk/aphil/movies/beef02.mov
sorenson3 Quicktime under 1mb

Gah. . . that teal background. I have always hated that default color. My eyes just start wiggen out. Its like framing your art in neon paint factory.

The facial shapes you have for this dude aren’t really working.

You have to really get a nice “eee”, “ooo”, “mmm”, “Ahh”, “Ffff” etc.

When you lay in your keyframes. . . you should never jolt from one extreme shape to the next. Blend them smoothly so that they overlap.

Also when your speaking. . . your lips make the shape of the sound before the sound comes out of your mouth. So in the case of the “FFFF” sound. . . your mouth makes that shape and is almost moving on to the “ire” part as the “f” sound is heard.

Hope that helps.

#45

Thanks for your time firstly guys! The feedback is great. Here are the mouthshapes separately -

that seems to be the main problem picked up.

  1. teal-background - the one and only time I tried changing AM’s colour preferences on my Mac it crashed so bad locking up my OS I wiped the hard disk and re-installed everything. I’ve had to learn to live with it. it works OK on my PC…

  2. creating the mouth shapes before the sound makes a LOT of sense - I actually found myself starting to do that but on “beef” only. I have re-worked a lot of the mouth shapes already and will try everything suggested.

  3. I think the tongue issue will be solved by putting up a final render or rendering from a choreography as the lighting isn’t picking up a tongue (it exists and is moving… honest!)

thanks again - I really appreciate your time on this guys

#46

Ok. Here is a head that I am working on. Be harsh.

Reminds me of a bust of Julius Caesar, especially with those eyeballs missing.:slight_smile: Anyway, this resemblance is my way of complimenting you. No unsightly creases as well, dare I say it, almost marble smooth! Are you using the new Bias normals or Porcelain? One suggestion, maybe the neck as it sweeps into the back of the cranium is too concaved?

Oh, and does anyone have any tips for making the inside of a mouth?

Apart from extruding back from the lips, no.

#47

Originally posted by alfiebabes
3) I think the tongue issue will be solved by putting up a final render or rendering from a choreography as the lighting isn’t picking up a tongue (it exists and is moving… honest!)

It’s not the existence of a tongue that’s the issue, but the fact that on those shapes that correspond to sounds made with the help of the tongue, the jaw is open too far.

To use examples from your pic, look at the “N-D-L-T” pose. The space between his teeth looks about a finger’s width. Put your finger between your teeth, then while holding that jaw position, touch your tongue to the roof of your mouth and try to make a “T” sound. The same jaw gap appears for “Th”, so try making that sound with your jaw open that wide as well. To me, neither one sounds very good. When I make those sounds normally, my jaw is hardly open at all, with very little gap between the upper and lower teeth, closer to what you’ve got for your “H-S-Z” pose. In fact, try holding your jaw that wide, and you can make a whole range of sounds: S-C-K-G-Z-N-D-T-L-Th.

I would suggest consolidating your poses. Make a generic “jaw open” pose (primarily using a bone) that opens the jaw really wide, and use that to control overall jaw movement. Open it wide, you’ve got a shape for Ahh…open it just a little, and you’ve got the beginnings of most of the tongue-dependent consonants: S-C-K-G-Z-N-D-T-L-Th.

Then make a series of lip-only poses that deform the lips and not the jaw (or with very little jaw movement, again using the bone so that things blend better) and use those in conjunction with the jaw pose to get the various shapes you need for representing the other sounds.

#48

Originally posted by John Keates
[B]Ok. Here is a head that I am working on. Be harsh.

Oh, and does anyone have any tips for making the inside of a mouth? Last time I felt like a drunken dentist with no training. [/B]

Nice head John. It´s seems that you’re using quite alot of splines there, but still it looks smooth. Maybe you could post a wireframe of it?
Only thing that bothers me is the ear, it just looks little bit out of place to me. It´s looks more creased than that head.

Keep up the good work!

#49

Hi, I’m also working on a new pic
LINK

#50

Originally posted by pequod
Reminds me of a bust of Julius Caesar, especially with those eyeballs missing.:slight_smile: Anyway, this resemblance is my way of complimenting you. No unsightly creases as well, dare I say it, almost marble smooth! Are you using the new Bias normals or Porcelain? One suggestion, maybe the neck as it sweeps into the back of the cranium is too concaved?

Hello there Stephen:wavey: Thanks for the comments. Yes, I am working on the eyes at the moment. I find it hard adding detail without the eyeball distorting. Maybe this is where the new “maintain curvature” thingy comes in.

I am using the new bias normals. Porcilain makes him even smoother the odd knoble here and there suits his face. Also, I am still getting problems with 5ppatches and porcilain. I hope they sort that out soon.

I am also using the new larger specular. It is realy good for skin (I think).

Cheers,
John

#51

Originally posted by daft
[B]Nice head John. It´s seems that you’re using quite alot of splines there, but still it looks smooth. Maybe you could post a wireframe of it?
Only thing that bothers me is the ear, it just looks little bit out of place to me. It´s looks more creased than that head.

Keep up the good work! [/B]

I hate doing ears! So much effort for something that people only notice when it looks wrong. If you have any further advice then it would be most welcome.

I am trying to attach another image file but I am getting a message saying that the file is too large. I have made it so small that you cant see it properly and put it on the lowest quality.

I may try again later.

Thanks,
John

#52

well,
i guess this is the place to post this. it is a WIP, but i also have a thread of it running too…so if it needs to be moved from here, i understand. anyway, i got a new book called “maya character animation” by jae-jin choi, and i’m finding it to be pretty useful with A:M too. he uses 5 main phonemes to cover the basics of lipsynch…so i tried out his method. here are the 5:

i think i might use the porcelin material to smooth out his face, but i will wait and add that at the very end…just in case it won’t be what i expect. i’m trying to stay motivated with this character, so far so good(the motivation… that is).

-kungfud0rk

#53

This is just a work in progress, the alleyway will have more building like objects in it soon, teh walls are just tiled right now.

http://www.polygondragon.com/random/alley_aa.jpg

The walls are modeled and the texture was whipped up in photoshop. The clouds are a digitally altered photograph i took with my digital camera.

-Mark Weller
AKA TacoBallZ
http://www.polygondragon.com

#54

Originally posted by kungfudork
well,
i guess this is the place to post this. it is a WIP, but i also have a thread of it running too…so if it needs to be moved from here, i understand. anyway, i got a new book called “maya character animation” by jae-jin choi, and i’m finding it to be pretty useful with A:M too. he uses 5 main phonemes to cover the basics of lipsynch…so i tried out his method. here are the 5:

Pretty good start on this, but there’s one missing from this group that kinda surprises me: an F/V shape. I’m not sure how the author of that other book can leave this out of a basic list of shapes. I use it often enough at work to know that it’s very essential.

While the shapes do look good, there seems to be some overlap that could be taken care of by consolidating shapes. Look at “Eh,” “E,” and “Ah.” They all have about the same mouth width, with the only major difference being how far the jaw is open. You could get the same basic effect by having two shapes: one that spreads the lips wider (side-to-side) than their default width, and one that just opens the jaw. Blend these two and you have all kinds of things at your disposal. Add a pursed “Ooo” shape (lips only) and you have several more you can use, again through blending w/ the others. Your current “ooo” shape looks more like a “ch” shape because the jaw is closed so far that the teeth are together. Take that basic lip shape and blend it with a slightly open jaw, and it would make for a better “ooo”. (for all I know, you could already be taking this approach; if so…um…way to go! :slight_smile: )

My main point is this: The movement of the jaw and the movement of the lips are completely separate things in real human speech, and for the best results, they should remain separate when preparing shapes for animation.

#55

jazbees: thanks for the comments, and i must agree about leaving out the “FV” phoneme…so i looked thru the book a bit more. i’m assuming what happened is that this book is a translation from chinese i think. i guess the FV isn’t a common pronunciation, not sure. SO anyway i went back and watched “Rigging a Face” by Raf Anzovin and decided i am gonna follow his method. So i hope in a few days to have a complete overhaul of the phonemes and facial expresions(time willing…i got a kungfu tournament in SF next weekend). I’m also trying to really clean up the face and smooth it out, so the character’s appeance might change a bit too.

-kungfudork:beer:

#56

OK, started building my model for my entry in this month’s CG-Talk’s animation challenge (I swear I’m going to finish, hell or high water!). It’s just the upper body so far. He’s going to be fairly cartoony and I really want to redo the hands… making them bigger. I’ll be updating as I get more finished.

Any feedback will be appreciated!

#57

Looking really nice. I would just add more detail around the upper chest area to show the bone a bit more. Great modeling. Will you combine this with more EI stuff?

#58

I myself kind of like the cartoony torso with the realistic looking arms.

Nice work so far.

#59

Thanks guys! I’m definitely trying to walk a balance between cartoony and realistic. I’ve done a few head drawing, but haven’t hit one I like yet. I think I’ll get the body nailed down first and hope that helps me design his head.

Here’s the latest. Tweaked the color to grey and added a lower body. Going to add some feet, probably tabi style shoes (split toes).

#60

hey MyFault:
i really like the look! i can’t wait to see how the head turns out! i did the same thing with my character, i started with a body and tried to build a head for it. it worked ok, but the head i built i really liked but it didn’t work on the body…so i guess i got two models out of one. just need to make a new body for the original head. anyway…i wanted to ask what settings you use for the colors. i want to get that soft look to my characters too. are you using just the regular surface controls or is it a material you constructed? could you please share your settings?

thanks,
kungfudork