Hybrid object constructions based on implicit and explicit representations

A system comprising:

• a memory component configured to store a mesh that models an exterior surface of a simulated object, and a neural representation of the exterior surface;
• and
• a processing device coupled to the memory component to perform operations that cause a requested modification to one or more surface regions of the mesh based on user input by:
• using a machine-learning model to incrementally update the surface regions of the mesh based on incremental changes applied by the machine-learning model to the neural representation for achieving the requested modification within corresponding portions of the exterior surface of the neural representation;
• and
• outputting the mesh for use in rendering the simulated object with the requested modification.

A non-transitory computer readable storage medium comprising instructions that, when executed, configure at least one processing device to:

• store a mesh that models an exterior surface of a simulated object, and a neural representation of the exterior surface;
• and
• execute a machine-learning model that causes a requested modification to one or more surface regions of the mesh based on user input by:
• incrementally update the surface regions of the mesh based on applying incremental changes to the neural representation for achieving the requested modification within corresponding portions of the exterior surface of the neural representation;
• and
• output the mesh for use in rendering the simulated object with the requested modification.