sprockets TV Commercial by Matt Campbell Greeting of Christmas Past by Gerry Mooney and Holmes Bryant! Learn to keyframe animate chains of bones. Gerald's 2024 Advent Calendar! The Snowman is coming! Realistic head model by Dan Skelton Vintage character and mo-cap animation by Joe Williamsen
sprockets
Recent Posts | Unread Content
Jump to content
Hash, Inc. - Animation:Master

Recommended Posts

Posted

Hello,

 

As hard as I've tried to come to terms with human body proportions (how long is the upper arm, lower arm, hand, upper leg, etc) I seem to still have issues! During the last stress test of model shown below I realized that things were not right so I placed it next to proper body proportions document (below).

 

Before I start scaling body parts this way and that way I wanted to get any feedback I could get from the forum. Also, for me, rigging starts with a 'guess' as to where the 'Body' bone goes (that's the bone that sticks out of the butt facing back wards). This is the pivot point of the legs and I'm never sure where it actually should be.

 

Image2.jpg Image3.jpg

Image4.jpg Image5.jpg

Image6.jpg

 

Any help will be greatly appreciated!

Rusty

Image1.jpg

  • Replies 16
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

  • Hash Fellow
Posted

I tried a few of those poses on myself and I'd say that the drawings are not necessarily accurate.

 

My knees don't reach even to my shoulders as in #3 and my back is not parallel to the floor as in #4.

Posted

I'd have to say the drawings are inaccurate as well.

 

Bicep/forearm roughly equal length from pivot to pivot (bicep/shoulder joint to elbow = elbow to forearm/wrist joint). Thigh/calf roughly equal length from pivot to pivot (thigh joint to knee = knee to ankle).

 

For the rest, I'd have to look into, but I would suggest using real reference photos, rather than drawings.

Posted

I agree... don't go by one source. I don't even think the artist who drew those references intended them to be blue prints. They are drawn in 2D by hand... you can fudge a lot of things. I bet if you took those drawings and carefully cut them apart and measured they wouldn't match up.

 

Also use the information rather than the visual reference. I've been in situations where something was "wrong" with the arms or legs. I had made some kind of mistake with the length of the limbs. I know where the hand should be at the sides, I know where the legs should be and the hips, I know the hand should be able to reach the shoulder... by using that information without a reference I was able to fix it by tweaking and measuring those distances.

 

--------

 

A very quick and easy way to check these things BEFORE rigging BEFORE adding any bones is to just use the mesh and the rotate tool. Select the points of say... the arm. Hit the "R" key and move the pivot to where the shoulder should be.

 

Now you can see instantly if the arm is long enough. You don't even need to rotate the points, just look at the rotation "ring" that indicates where the arm would be if rotated to the side. Do the same with the forearm and upper arm. I even put in markers and draw two point splines to quickly and accurately gauge the position and proportions of limbs before doing any rigging or even adding bones.

 

If you want you can also rotate the mesh and then undo it. I've done that before. Rotate the arm at the shoulder and the elbow to see how it looks and then hit undo 3 times or whatever.

 

-vern

Posted

Bodies aren't all the same. Art students learn rules of thumb for average bodies first. Once you know the numbers, you can view any actual body as a specificic departure. Like, maybe Robcat is Greek - just a guess - Greeks tend to have longer torsos than legs.

 

A wooden mannequin might be a good investment. You can get one at any art supply store or Barnes & Noble. The really cheap ones aren't so hot; try it out before buying to see that it will take and hold extreme poses. The one I got won't put its arm across its chest; I'm going to have to buy another.

 

You've probably seen this:

body.jpg

Posted

The hands look too small proportion wise. If you put your hand in front of your face, and the base of your hand on your chin, your middle finger should reach the edge of the start of your hairline (or more than half of your forehead in the absence of hairline). The hand in your character looks like it can barely reaches the eye line. And to be sure, the proportion of the hand should cover from the heel to the bunion of the foot.

 

I think once you get that done it will be sufficient to cover up any other glaring misproportions.

Posted
If you put your hand in front of your face, and the base of your hand on your chin, your middle finger should reach the edge of the start of your hairline (or more than half of your forehead in the absence of hairline).

 

Glad you clarified that...

 

I was thinking my hand needed to be an extra 3 or 4 inches long... like some mutant E.T. or schwa. My middle finger would need to be HUGE to reach to the back of my head.

 

;)

 

-vern

Posted

Thanks for all the responses and all the info! The source used in my posting was actually the only one that I hadn't used before rigging. The examples I used from it, to the extent I could tell, matched my own body.

 

BTW, I litterally have every source I could find on the Internet -- enough to make you crazy because they do not all agree.

Image2.jpg

 

Image1.jpg

 

Mostly I used myself to double check the sources I found that did agree (for the most part) but, even doing this there were inconsistencies. Believe it or not, even if I try go strictly by my own body I run into problems and things don't look right! I think its a case of trying to digitally measure an analog system (no pun intended). Meet one requirement and another fails... plus, you tend to lose your 'control' or starting reference point (you have to have one part you say is correct and use this as a starting point). I finally settled on what I thought was the best combination, checked this against myself as well as other people I had pictures of and a few months ago I posted this as a guide to average body proportions. At the end of the day everyone is a little different.

 

Despite my pre-rigging efforts, somehow I still managed to mess up LOL. I believe:

a) The hands are too small and need about a 120% increase (and once I increase these by 120%, other parts which use the hands as a basis will be off). Granting that my reference is a little accurate, image one shows this.

B) Image 2 test correct on everyone in my family and clearly indicates that either a) the upper arm is too short, or B) the lower arm is too long or, c) a little of both.

c) Image 3 indicates that the upper body is too long... or, the lower body is too short... or, a little of each. Perhaps the image is not 100% but my model is way off.

d) Looking at image 5 my gut feeling is that the body bone (or upper leg pivot) is a little too low (which will impact item c to some extent)

 

For now I am discounting image 4 but will come back to it after the above changes are made.

 

I'm surprised so many of you said that the reference I used was incorrect. It is a guide for 2D artists but I can't see how this matters. After what I've seen, I have to take all of the references I found with a grain of salt -- still, if something is 'way' off, I have to review it.

 

Finally, since I'll be adjusting mutually exclusive systems using the root bone for that system and holding down the key to expand the mesh as well, I don't think the rig, fan bones, CP weights or Smart Skin will be affected in most cases.

 

Rusty

Posted

At some point you have to... go solo... you have to let go of the reference and... use the artistic side of the brain.

 

A problem with photo references at least is camera distortion. It can be subtle but a "real" camera is not like the display in AM. AM is "flat" everything is the same scale no matter how you rotate. With photo reference even the slightest rotation completely changes the reference points... they will never line up. Not as big a problem with body proportions I suppose because they are closer to the same position. Another problem is slight changes in angle from one reference to the next. A slight tilt forward can completely mess up reference points.

 

Another problem is the rotation pivot points of the bones or limbs. If after creating a model based on reference or lengths of limbs and you put the rotation of a joint in the wrong spot it won't be correct. Finding the proper pivot point is crucial.

 

Yet another thing is what the body does when it rotates. I noticed from the original images that in the drawings there is all kinds of bending and shifting of different body parts. These small things can change how the parts relate to each other. If the back bends a little when the character bends over, the back gets shorter. If your model doesn't have a gradual bend of all the bones in the back it won't match the reference.

 

From what I could see you are pretty darn close... a bit of tweaking and you are almost there.

 

-vern

  • Admin
Posted

I think to a very good degree this illustrates why animators adopted 'Squash and Stretch' and 'Exaggeration' as fundamentals with regard to animated bodies. This is no doubt why more cartoons have been used than realistic humans over the years. You can get away with more with squash and stretch in imaginary characters.

 

Vern hits on an excellent point with the camera's view/perspective.

For some poses we may need to break the model's rig in order to get the proper effect of foreshortening.

We may often find poses where 'reality' doesn't look right or where something not filmed realistically looks somehow... better.

 

I think this is why I am attracted to the cartoon form over more realistic rendering/porportions.

In its own way the exaggeration makes a lot more sense to my brain... with stylized exaggeration I immediately accept that what I'm seeing is 'real enough' and I suspend my belief and move forward into the story. Going too realistic reminds the brain that what it is seeing is created with a computer and that can break the storyteller's spell.

 

The roto you've placed next to the character has considerable exaggeration however even for a line drawing. That first one has very 'superheroic' porportions. In order to get closer to that (I think) you would have to break with realism. (go with the flow.., chart your own course... break the mould)

 

From what I've seen of your characters thus far I can see you've strived to reach a good degree of realism in your work and I'd say you've achieved considerable success there. Keep it up!

Posted

The reference drawings are wrong that's for sure. Maybe you could join a life drawing club and other than being an excellent way to develop skills, very often the models don't mind if you take a few photos. The way to get around distortion is to take the picture from a distance and zoom in. This flattens the image.

Posted
At some point you have to... go solo... you have to let go of the reference and... use the artistic side of the brain.

 

A problem with photo references at least is camera distortion. It can be subtle but a "real" camera is not like the display in AM. AM is "flat" everything is the same scale no matter how you rotate. With photo reference even the slightest rotation completely changes the reference points... they will never line up. Not as big a problem with body proportions I suppose because they are closer to the same position. Another problem is slight changes in angle from one reference to the next. A slight tilt forward can completely mess up reference points.

 

Another problem is the rotation pivot points of the bones or limbs. If after creating a model based on reference or lengths of limbs and you put the rotation of a joint in the wrong spot it won't be correct. Finding the proper pivot point is crucial.

 

Yet another thing is what the body does when it rotates. I noticed from the original images that in the drawings there is all kinds of bending and shifting of different body parts. These small things can change how the parts relate to each other. If the back bends a little when the character bends over, the back gets shorter. If your model doesn't have a gradual bend of all the bones in the back it won't match the reference.

 

From what I could see you are pretty darn close... a bit of tweaking and you are almost there.

 

-vern

 

You've hit the nail on the head with all points Vern. I thought I was at the point were I could 'just do it' but the virtual actress model Monique who appeared in the Horrorthon trailer knocked me for a loop -- well after completion I noticed that her forearms were 'too long'. Instead of making any kind of physical correction at that point in the Horrorthon project I hid this defect with camera angles and scene staging. I then became determined not to make that mistake again and I'm now perhaps extra sensitive to this.

 

r

Posted
I think to a very good degree this illustrates why animators adopted 'Squash and Stretch' and 'Exaggeration' as fundamentals with regard to animated bodies. This is no doubt why more cartoons have been used than realistic humans over the years. You can get away with more with squash and stretch in imaginary characters.

 

Vern hits on an excellent point with the camera's view/perspective.

For some poses we may need to break the model's rig in order to get the proper effect of foreshortening.

We may often find poses where 'reality' doesn't look right or where something not filmed realistically looks somehow... better.

 

I think this is why I am attracted to the cartoon form over more realistic rendering/porportions.

In its own way the exaggeration makes a lot more sense to my brain... with stylized exaggeration I immediately accept that what I'm seeing is 'real enough' and I suspend my belief and move forward into the story. Going too realistic reminds the brain that what it is seeing is created with a computer and that can break the storyteller's spell.

 

The roto you've placed next to the character has considerable exaggeration however even for a line drawing. That first one has very 'superheroic' porportions. In order to get closer to that (I think) you would have to break with realism. (go with the flow.., chart your own course... break the mould)

 

From what I've seen of your characters thus far I can see you've strived to reach a good degree of realism in your work and I'd say you've achieved considerable success there. Keep it up!

 

Thank you for the kind words Rodney. My turn of the century 'million dollar idea' to produce animated book trailers has been my singular goal these last eight years. I have not IMHO reached a marketable artistic state yet plus I've had to face the fact that although I may achieve the technical and artistic level needed eventually, I don't know how to do 'the rest' (take it from pipe dream to reality... get investors, start a company, sell the idea, protect personal exposure, etc.). There's also the fact that my wife and I are comfortably retired, we don't need money and I'm not really anxious to return to the realm of deadlines and stress (not hungry).

 

But, LOL, I digress... I'm still pushing forward and its fun if nothing else.

 

r

Posted
The reference drawings are wrong that's for sure. Maybe you could join a life drawing club and other than being an excellent way to develop skills, very often the models don't mind if you take a few photos. The way to get around distortion is to take the picture from a distance and zoom in. This flattens the image.

 

This of course is the proper course and in a way I have... I just have to make the time for it (between writing and animation). But I'm also an engineer by nature and reducing things to digital formulas and structures, if not really 'art', is always a desire, a trait and even a habit. I do need to get back to my sketching tutor and the lessons we started -- I very much enjoy it.

 

r

Posted

Re: camera distortion, keep in mind that your eye distorts for the same reason. For your roto, you want a shot taken with the camera at a great distance so you can build an objectively correct model. But then you have to render it in a Chor with carefully chosen camera placement to see how it really looks. Shaded renders in the modelling window won't look right because our own eyes are used to seeing things with perspective distortion.

Posted
Re: camera distortion, keep in mind that your eye distorts for the same reason. For your roto, you want a shot taken with the camera at a great distance so you can build an objectively correct model. But then you have to render it in a Chor with carefully chosen camera placement to see how it really looks. Shaded renders in the modelling window won't look right because our own eyes are used to seeing things with perspective distortion.

 

Very good point. Thanks!

 

Rusty

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...