LSD-StructureNet: Modeling Levels of Structural Detail in 3D Part Hierarchies
This addresses a practical limitation for applications like 3D CAD design that require adjustments at multiple levels of detail.
The paper tackles the problem of existing 3D shape generative models being unable to conditionally sample individual parts without modifying the entire shape, and introduces LSD-StructureNet, which enables re-generation of parts at arbitrary hierarchy depths without impacting inference speed or output realism and diversity.
Generative models for 3D shapes represented by hierarchies of parts can generate realistic and diverse sets of outputs. However, existing models suffer from the key practical limitation of modelling shapes holistically and thus cannot perform conditional sampling, i.e. they are not able to generate variants on individual parts of generated shapes without modifying the rest of the shape. This is limiting for applications such as 3D CAD design that involve adjusting created shapes at multiple levels of detail. To address this, we introduce LSD-StructureNet, an augmentation to the StructureNet architecture that enables re-generation of parts situated at arbitrary positions in the hierarchies of its outputs. We achieve this by learning individual, probabilistic conditional decoders for each hierarchy depth. We evaluate LSD-StructureNet on the PartNet dataset, the largest dataset of 3D shapes represented by hierarchies of parts. Our results show that contrarily to existing methods, LSD-StructureNet can perform conditional sampling without impacting inference speed or the realism and diversity of its outputs.