Hi everyone,

I’m having trouble achieving the expected muscle definition using OTIS in Houdini.

For some reason, I’m not able to get the same level of visible muscle detail that I can achieve with Vellum (please see the attached reference images). In terms of realistic muscle deformation, the results from OTIS appear much flatter and resemble something that could easily be achieved with a simple blendshape, rather than exhibiting natural muscle bulging and volume shifts.

Additionally, I’ve noticed that OTIS does not seem to produce realistic muscle behavior similar to Ziva, where muscles naturally bulge and shift upward during contraction. Instead, the deformation feels limited and lacks dynamic volume change.

I’ve attached my Houdini file (Version 21.0.631) in case anyone would be willing to take a look and help me understand what I might be missing or setting up incorrectly that’s preventing proper muscle definition with OTIS.

Any guidance would be greatly appreciated. Thanks in advance!

Hi

I took a look at your file, and I can see why you're disappointed with the results! There are definitely things one can do to improve it.

To begin with, the muscles aren't activating properly. When I copied the tension line and muscle flex settings from the original Vellum setup, I got nice activation and bulging in the Otis simulation. There are a couple of reasons why the original setup gives better activation results than the new one you have with default settings.

Firstly, the new activation workflow is not expected to give correct activation with the default settings - you are expected to tweak the settings in the "Muscle Tension Lines Activate" SOP to get the expected activation. This is even more true the further the muscle ends are from their correct bone attachment points, as is the case with this model (and its muscleendmask), compared to Otto, for instance, because it is less likely to capture the "true" length shortening of the muscle with bone movement.

Secondly, the original setup has a couple of tension lines drawn that don't correspond to any specific muscle, and these are driving most of the muscles. Even though the activation ratios for these lines are at their default values, there was clearly quite a bit of thought put into where the lines should be drawn and which muscles to connect to which lines. This workflow is still valid and supported for those who prefer to work with fewer lines. It is a matter of personal preference. You can also use a combined method where you use the auto tension lines for most of your muscles, but then merge in some manually drawn tension lines for certain muscles. Legs are often areas that tend to benefit from custom lines because muscle activation isn't fully driven by length changes alone, since the force interaction with the ground adds another component to the activation. I demonstrate how to do this in my new Masterclass which should be released within the next few days.

gregjenings
Additionally, I’ve noticed that OTIS does not seem to produce realistic muscle behavior similar to Ziva, where muscles naturally bulge and shift upward during contraction. Instead, the deformation feels limited and lacks dynamic volume change.

I'm not sure if you mean for this file specifically, in which case I agree, and the main issue was because you didn't have proper activation, or more generally. If you mean more generally I would love to see examples to back this up, as I have limited experience with Ziva.

One way to control how the muscle contracts is with the tendon mask. If you want the muscle to contract more towards the top of the muscle, then make the bottom tendon mask longer than the top tendon mask (using the tendon mask radius parameter in Muscle Properties Otis).

As for muscle definition, there is already a lot more of this once the muscle activation is fixed. To push it even further though, you have various options.

Because the scale of this character is so big, I recommend increasing the Tissue Muscle Attachment distance threshold in the Tissue Constraint Properties Otis node to something like 0.05. The Tissue Muscle attachments are the best way to get muscle definition without making the character look more lean than it was modelled originally.

Furthermore, the Vellum example you are comparing with achieved the muscle definition by shortening the rest length for the tissue simulation. This was the only way to really achieve muscle definition in Vellum. In Otis it is less necessary, because we have Tissue-Muscle constraints, but if you want to get a look that is similar to the vellum result, decreasing the rest length scale is necessary on the tissue. I've attached a screenshot with and without tissue rest scale shrinkage).

Some settings that I changed:
Tissue core damping ratio: increased multiplier from 1e-6 to 1e-4
Tissue solid layer shape stiffness: decreased from 10 000 to 5000.
Tissue Muscle Attachment Stiffness: increased from 1000 to 5000, distance threshold increased from 0.01 to 0.05.
General Muscle Shape stiffness: decreased from 100 000 to 50 000.
Muscle Volume stiffness: decreased from 150 to 25.
General Fiber Strength: decreased from 20 to 10.
Fiber scale range 0 to 1
Tendon Mask radius 0.2
Muscle Ends Stiffness: increased from 1500 to 15 000.

Various per muscle overrides to the fiber strength, fiber scale range, and tendon mask radius based on the settings in the original vellum setup (RMC on tab in original node-> copy instance, RMC on tab in new node-> paste instance).

Painted the top edge of the gluteus medius to have a muscle end mask.

If you are using the rest length scale, make sure to key in the rest blend like I did in the file.

I hope that helps!

I forgot to add that I unlocked the "Otis Muscle and Tissue Configure" node and tweaked the point deform settings for the "Deform to Bones on Reference Frame" option, as they weren't working properly on this large a character. I'll try to come up with a solution for this soon that doesn't change the behaviour for everyone. With the default settings it looks a bit weird on the first frame of the simulation. It's not too bad an issue, as the solver uses the stored tpose to fix it by the second frame, but just be careful to not use the first frame as the "rest" frame for any skin deformations or the like. I also added "tpose" to the output attributes on the "Otis Muscle and Tissue Configure" node so that the tpose can be extracted from the sim.

hi Liesbeth : thanks for the detailed writeup! Can you explain the difference between muscle ends and tendon mask ? when and where would you use muscle ends Vs Tendon mask ?

thanks

carthick
hi Liesbeth : thanks for the detailed writeup! Can you explain the difference between muscle ends and tendon mask ? when and where would you use muscle ends Vs Tendon mask ?

I explain it in the first two minutes of this video (the rest of the video is relevant to Vellum only, not to Otis) https://www.youtube.com/watch?v=rmWmYfKfO5U&list=PLXNFA1EysfYlEaZ7w6YbQpWqZY6-l2Onu&index=6 [www.youtube.com]

Alternatively you can watch it here https://www.sidefx.com/tutorials/muscles-masterclass-essentials/ [www.sidefx.com] where it is the sixth video.