Houdini 22.0 Machine Learning Trainable ML Solutions

Animation and character nodes for trainable solutions

ML Deform

This set of nodes allow a character skin to be deformed using quasistatic tissue simulation. Poses can be generated in parallel because there are no time dependencies between the simulations.

Name

Type

Description

Armature Deform SOP and Armature Capture SOP

Data Set Processing

Deform organic tissue repeatedly using skeleton poses. You can use these nodes to generate training data for an ML deformer and other applications outside of ML.

ML Train Deformer (Recipe)

Model training

The ML Train Deformer recipe creates a setup with two entities for which ML Deformer models can be trained.

ML Deform SOP

Inference

Uses a trained ML model to deform the skin of a character. It uses the same inputs as Bone Deform SOP with a fourth input for residual blend shapes. Each residual blend shape defines a displacement point in the skin mesh. Internally, the ML model predicts weights to combine the residual blend shapes to apply a correction to the rest skin.

APEX Add ML Deformer SOP

Inference

Allows a model trained using the ML Train Deformer recipe to be used in APEX rig.

Agent Add ML Deformer SOP

Inference

Adds shapes to an agent and configures them to use a trained ML model which deforms the skin of a character. Internally, the ML model predicts weights to combine the residual blend shapes to apply a correction to the rest skin geometry.

Trainable ML Solutions

ML recipes

ML workflows

ML nodes in other areas