• Hi Guest!

    Please be aware that we have released a critical security patch for VaM. We strongly recommend updating to version 1.22.0.7 using the VaM_Updater found in your installation folder.

    Details about the security patch can be found here.
Demos or "Lite" versions of Paid Content
changelog v23
Features
  • Added two new morphs to adjust arms, created by the one and only CheesyFX.
    They allow scaling without affecting the elbow angle.
  • Added a UI slider to rotate both females and the drawn skeleton.
  • You can pick a file name when saving and loading ?
Changes
  • Reverted the limits for the head big and body size morphs to their VaM defaults.
  • Added sliders to adjust said limits in the plugin's config (scene plugin).
    Most users should be OK with the defaults but if your models have large size differences this will help.
General Improvements
  • Improved automatic estimation of body size morph values. (the body's volume)
  • Improved loss calculation to prioritise outliers.
  • Added the elbow's angle to the network to improve accuracy of arm alignment.
Advanced stuff
  • Implemented network pruning to reduce processing time.
    Probably don't mess with this yourself unless you already know what that even means.
changelog v22
Fixes
  • Fixed the UI not being anchored to the camera correctly.
General improvements
  • Increased the min/max values for morphs from -1/1 to -2/2.
    This affects existing training data but should be compensated for relatively quickly.
  • Changed how the training rate is calculated and configured.
    You can adjust this in the session plugin's settings if you wish to.
    • increase to speed up processing a single look
    • decrease to prevent overfitting while training on a bunch of morph presets
changelog v21
General improvements
  • Removed the Hip up-down morph because it had too much of a visual impact.
    You will have to re-train your saved networks, sorry.
  • All output morphs will now be zeroed before applying/training.
    • Pressing "Apply" multiple times will always have the same outcome.
    • Letting training go on for multiple runs will yield better and more consistent results.
  • Body-to-head ratio adjustments will now scale "Body Size" in addition to "Head Big".
    This should prevent height differences to result in a stretched model with a big head or squeezed with a small head.
UI
  • Added an overlay view highlighting differences between models.
  • The UI is now locked to your PoV and moves with camera.
  • Added a toggle to enable/disable automatic head scaling.
  • Added a toggle to enable/disable the UI.
Fixes
  • Unclamped the head scale morph (Head Big). Sometimes -1 to 1 just wasn't enough range.

Showcase

sneak peek at v21

added the following morphs from Morphs.Morphs Mega Pack.1
  • Hip UpDown
  • Shin Length
  • Thigh Apart
  • Thigh Length
  • Waist FrontBack
  • Waist Height
Let me know if you see an improvement or run into issues.
the ratio between body height and head height will be kept constant now
The required physics settings "Freeze Physics", "Soft Body Physics" and "Glute Soft Physics Enabled" will now be configured automatically before applying or training.
The bundled morph for arm length sometimes didn't load.
I updated the plugin page, too. Hopefully, it's a bit easier to understand what this actually does and what it's for.
  • Improved UI
  • Fresh training data which will yield far better results when you hit "Apply"
  • Reduced training rate to prevent overfitting
  • Reduced the network depth cutting the time each generation takes roughly in half
  • Training rate can be adjusted in the plugin's options (scene plugin)
    • Decrease training rate when you want to teach the network a bunch of different morphs, thereby improving its versatility and accuracy.
    • Increase training rate if you want to quickly adapt to a specific morph. This will overfit the model, reducing its accuracy when applied to other morphs. Be aware if you save the network.
fixed a major bug with training data persistence and re-trained the network
Back
Top Bottom