Modeling secondary motion based on three-dimensional models

A system comprising:

• a memory component;
• and
• a processing device coupled to the memory component, the processing device to perform operations including:
• receiving a plurality of three-dimensional object models representing an object in primary motion;
• encoding, using one or more machine learning models, surface normals and velocities of corresponding portions of a particular three-dimensional object model based on the plurality of three-dimensional object models;
• synthesizing, using the one or more machine learning models, a two-dimensional silhouette of at least one three-dimensional feature to which the particular three-dimensional object model is physically connected based on the surface normals and the velocities, the two-dimensional silhouette subjected to secondary motion that is different from and induced by the primary motion;
• modeling, using the one or more machine learning models, the at least one three-dimensional feature subjected to the secondary motion based on the surface normals, the velocities, and the two-dimensional silhouette;
• and
• rendering the particular three-dimensional object model having the at least one three-dimensional feature.

A non-transitory computer-readable medium storing executable instructions, which when executed by a processing device, cause the processing device to perform operations comprising:

• receiving three-dimensional motion descriptors relating to a particular three-dimensional object model of an object in primary motion;
• receiving at least one feature that is subject to secondary motion to be applied to the particular three-dimensional object model;
• synthesizing a two-dimensional silhouette of the at least one feature modeled as physically connected to the particular three-dimensional object model and subjected to the secondary motion based on the three-dimensional motion descriptors, the secondary motion being different from and induced by the primary motion;
• determining surface normals of the at least one feature subjected to the secondary motion based on the three-dimensional motion descriptors and the two-dimensional silhouette;
• and
• modeling the at least one feature subjected to the secondary motion by combining the two-dimensional silhouette and the surface normals.